Changelog

2026-10-25

Keeping one entry leaves one line on the sheet. When the breakdown joined two entries for one thing that two departments had both listed, the joined-away wording could stay on the second department's list beside the one it kept, so the sheet showed the same thing twice while the card said it was one. Now the join takes it off every department's list on those scenes, unless another entry goes by that name, and the departments' check no longer adds the joined-away wording back on a scene the kept entry is already on. Undo puts both lines back where they were.
A cast card says what's on the sheet after you undo it. When you pressed Undo on a person the breakdown added to the cast, or on two names it joined into one role, the card under Saved answers still said the person was on the cast list, or the joined role was on every scene, with the breakdown's date under it. Now the card says what you did ("Took … off the cast list", or "Put … back apart as two roles"), who did it and when, and "On the sheet now" and the names at the top of the card show the cast list as it stands: each role on its own scenes, or the person off the list. Undo under Saved answers puts the card back exactly as it was.
Answered for you shows the same number as the finish line. The count beside Answered for you, and on Accept all, could say 10 where the finish line, Read history and HQ said 18: it skipped cast records, like a person the breakdown added or two names it joined into one role, and counted a card with two changes twice. It now counts each card once, cast records included. And when you press Undo on a cast record, you can now undo that too: Undo under Saved answers puts the person or the join back, and the card goes back under Answered for you as it was.
Read history and the finish line count what your tray counts. After a big read, Read history could say 296 waiting for you while the tray showed 125, because it counted every card before cards asking the same question were grouped. Read history now says two numbers, how many are waiting for you and how many were answered for you with Undo, the same two the tray shows, and they change as you answer.
A department choice names only your show's departments, and says why it recommends one. A card asking which department should own something no longer offers or recommends a department your show books nothing under, like Casting, and a department it named by another word (Picture Vehicles for your Picture Cars) now shows once, under your show's name, and files there. When a card asks once for an item filed by several departments, the department it recommends says how many of the item's entries the breakdown suggested it for, and a card that recommends nothing says so instead of leaving you to wonder. Two cards asking which department owns one item on one scene are now one card, so one answer files both. A card asking about a department with no department in mind is asked as the question it really is, whether the item is already on the list. A question about an item that already has something answered for you now shows that answer on the same card, with its own Keep and Undo, instead of on a second card that could disagree. Keep there saves your answer with its own Undo, Accept all keeps those too, and the count beside Answered for you includes them, with a line saying how many are shown on questions above. And a card that joined two spellings of one person, written as "B is the same person as A", now offers Undo, which puts B back as their own role on the scenes B is written in.
A card says what it's really asking, and why. When the script check sends back a join the read made because it thinks the two are different things, the card now puts Undo forward and says why; it used to put Keep forward, against the check. When the check only thinks it's your call, nothing is put forward. A card the read filed under both costume and props, which has nothing to undo, stays under Answered for you with the check's doubt on it, instead of coming to you as a question with one answer. A casting call says casting is your call, not that the line couldn't be found. A card asking whether entries are one item names every scene they sit on, each a link. And when the script check finds an item on a department's list is already covered by another entry, the card says which entry covers it, on which list and scene, instead of saying the item is on that department's list, or, once you keep it, that it came off the sheet.
A part the read filed itself no longer keeps a question on your tray. When the read joined two entries on one scene and the script check then found they are really two things, the card still came to you if the read had also listed another department's part on that item, as if someone's work rested on it. Now only a part a person or a department filed counts. The read undoes the join itself, both entries stand on the sheet, and the card waits under Answered for you, where Ask me this joins them again and puts the question back. A part you or a department named, changed or took off still keeps the question with you.
Keep on anything answered for you can be undone. Pressing Keep on a card under Answered for you closed it with no way back, whether the breakdown or the script check had answered it. Now the kept card shows Undo straight away, and again under Saved answers, and Undo puts it back under Answered for you exactly as it was. Keep never moves anything on the sheets, so neither does Undo. After Undo on Put the join back, the card no longer shows an old saved answer. Accept all works the same way: each card it keeps shows under Saved answers with its own Undo, and undoing one leaves the others kept. A department's list of items the script check answered works the same way, one item at a time: each Keep has its own Undo, which puts the card back under Answered for you, not back to you as a question. A card answered for you by undoing a join now names both entries, with links, and says they were put back apart, instead of offering to keep one.
You can put back a join the script check undid. When the script check decided two entries the read had joined are really two things, it split them and closed the card, and there was no way to join them again from the tray. Now the split waits under Answered for you, naming both entries with links to each, with Keep and Put the join back. Put the join back makes them one entry again, exactly as the read had it, and saves that as your answer. Undo on your answer splits them again and puts the check's card back, so you can change your mind as often as you like. Read again keeps whichever answer stands.

2026-10-24

Every way a writer cues a line counts as that person speaking. A cue with two marks like (V.O.) (CONT'D), a dual-dialogue cue, a numbered role like #2, a name with an accent or an "Mc", and two names sharing one cue are now read as the character's own line, the same as a plain cue. Before, the breakdown could miss that the person speaks there and treat them as only mentioned. A line of action in capitals is still action.
Read again keeps your answers about people. If someone was booked as background, by you or by the script check, a later Read again, or a new draft with them on the same lines, used to put them back on the cast list and take them off the scene's background without asking. Now they stay background, including in a new scene that only mentions them. If the new draft gives them a line in a scene the answer never covered, or someone booked them as cast in another scene, you get one card and nobody is booked until it's answered. Two names you said were the same person stay one person, and someone you booked stays booked. A child you said needs more than one performer keeps one booking per age, and you aren't asked again whether the ages are the same person.
Cards that only repeat a question are answered for you, and say why. At the end of every read, and after Check them now, the read looks over the cards still waiting and answers the ones that ask nothing new: two entries on the very same scenes, a name said once, a join the read made on one scene, a second question on an item another card already asks, a card with only one answer. Each sits under Answered for you with one line saying why, and nothing on the sheets changes. Keep closes it; Ask me this puts the card back on the tray as it was. Two entries on one scene, or a pair one line of the script names, are answered this way even when the item check asked whether they're one build or two. When the read joined two entries on one scene and the script check says they're two things, the join is undone so both stay as the script has them; Ask me this joins them again. Casting calls, booking calls, department choices with a real pick, and one-build-or-two choices across scenes are never answered this way. The finish line and Read history say how many were answered and why, so each kind can be fixed at its source.
Read again finishes when its check answers someone the first read asked about. If the first read left a person as a question and the check on a re-read settled them as background, the whole check stopped and the Script page said questions still needed checking, with nothing answered. Now the check finishes, the card is answered for you, and the person sits on the scene's background.
A lead stays a lead on the scene sheets. When a card asking whether two spellings are one role was answered, by you or by the script check, every scene that character was in listed them as a day player, and the scenes the answer added did too, while the cast list still had them as a principal or a voice. The strip board and Day Out of Days showed the same mistake. Now each scene keeps the character's own role, and Undo on the answer puts the scenes back as they were.
Casting calls and one-build-or-two choices stay with you. The script check no longer answers a card for you when it's a casting call (one part the script writes at more than one age) or a choice the item check marked as one build or two, however sure it is and whatever line it quotes. Those cards stay with you, with the check's answer put forward and the reason beside it. An answer with no line to stand on still needs the check to be at least 80% sure.
Tell us when a card shouldn't be a question. Every question on your tray now has a tick box, "This shouldn't be a question". Tick it, say why, and the card comes off your list. It isn't answered, nothing on the board changes, and anyone else can still answer it. It waits under Set aside by you with your reason, and Undo puts it back. We read every reason and use them to decide which questions the breakdown should stop asking.

2026-10-23

The script check stops asking what the read already settled. When the check doubts a join the read made of two entries on the very same scenes, the join stays under Answered for you with Undo, the way the read leaves any same-scene pair. When it doubts a department's part on an item another card already asks about, the part rides that card, is listed on it, and your answer there covers it; put that answer back and the part goes back on the card. A split whose own line names both pieces stays made, with Undo.
Undo works on a join the check sends back to you. When the script check thinks the read joined two entries that are really two things, the card comes back to you with Keep and Undo. Undo used to say the item had changed even though nobody had touched it, because the read had gone on to fill in the scene's cast and background afterwards. Now Undo puts both entries back on the sheet and keeps everything else the read wrote on those scenes. If someone has edited the scene by hand since the read, Undo still stops and asks you to look first.

2026-10-22

Fewer cards, each with its line. The script check now reads the whole script and every card in your tray at once, the way a line producer would, and asks you only where two answers are defensible and something rests on the choice: how many to build, who owns it, how it's made, who to book. Everything else is answered for you with the line from the script, or from the show's brief, that answers it, and Undo. When the check thinks the script answers a card but can't find the line it quoted, or the brief only leans one way, the card stays with you, with that answer put forward and the reason beside it.
A cast row on the schedule table reads clean. The line under each name in the Day Out of Days is the counts and the span, like "2 work · 14 idle · 16 days · 31 Jul – 23 Aug". The imported report's day-by-day notes no longer pile into it; they sit on the day itself. The header says "No shoot today" on a day off instead of "Day —".
Two cast numbers can swap. On the Cast page, type the new numbers in as many rows as you like, then press Save the numbers. They save together, so Dek can take Mara's 1 while Mara takes Dek's 4. If two rows hold one number as you type, the page says which one, and the warning goes away the moment you fix it. Nothing is saved until you press Save, and a refusal saves nothing. The board and the table show the new numbers right away.
Inside the strip board, Schedule takes you back to the show's schedule. It used to land on an empty page that told you to open the schedule from the show.
Dragging a strip between two strips holds still. On the strip board, a strip carried to a place between two others now shows one steady "Place scene … here" slot for as long as you hold it there, and drops where the slot is. Before, the slot could jump to the top of the day and back every frame, so the strips beside it shook.
A scene number comes with the page it's on. When a script arrives without scene numbers, the breakdown numbers the scenes itself, so "scene 34" is a number the writer never printed. Now every card, quote, scene list, scene page and finish line on that show says "scene 34 (page 31)", using the page number printed in the draft where the scene starts, never the PDF's own page count, so a title page and front matter don't throw it off. A page that prints no number says "no page number". A script that prints its own scene numbers reads exactly as before.
One question for one pair of items. When the breakdown asks whether one item is already on the list as another, and also asks whether the same two are the same thing, you now get one card, not two. The other is closed and says which card asks it, and your answer on that card settles both.
Every entry on a card is one you can find in the script. A card asking whether two things are the same, or whether two names are one person, now lists each entry with its scenes. An entry the board has on no scene takes its scenes from where it came from: where the last read left it, the scene the departments' check placed it in, the lines an earlier read heard someone on, or the line that writes a name. An entry nothing places is never put to you and never joined to anything: it's kept apart, with a note under Answered for you saying why, and Undo closes the note. A quote left over from an earlier read with no scene now shows the scene its line is in.
Fewer cards the script doesn't call for. The script read asks whether two entries are the same thing only when a line of the script names both, and asks whether an entry is two things only when the line it quotes names each part. When the read both places an item on another scene and joins another wording into it, you see one record with an Undo for each change. The read only puts an item on another scene when that scene's line names it, so it never asks you whether an item is in a scene. When it isn't clear which department owns something, you're asked to choose between the two departments it could be, with the line that names the item, instead of picking from every department on the show; when the same choice comes up for several items, it's asked once, naming each item and its scene, your answer covers them all, and one Put it back puts them all back. Each item gets one such choice, with any change the read made to that item on the same card, and a choice between two art departments other than props, like scenic or set decoration against production design, is the art department's own and isn't asked. What the departments' second look adds is checked the same way as the read's own list, with any item of another department on the same scene that could be the same thing beside it. Something the script names stays on the list even when the read first took it as part of something else, and the departments' second look lists everything the page names again, including each person's look and every place a scene happens.
The script read writes down what the page needs and leaves the calls to you. When the read sees a silent person fall, it puts them on the cast list with the line, and leaves whether it's a stunt, and how they're booked, to you. A younger self is its own part, and the read never assumes a second actor; if you've said one actor plays both, it reads that as written. A dummy is something to build, not a cast member. Rain on a window or a car passing outside is listed as what must be seen, with how it's made left open. Two phones that belong to two people stay two phones. When the script doesn't say, the read says so instead of guessing, and its checks come back empty when there's nothing wrong.
The script read sees the page as it's laid out. A PDF script now keeps where each line sits on the page, so the read can tell a character's name from a prop in capitals, a parenthetical from an action line in brackets, and a transition from something in the scene, the way you do. A Final Draft file keeps each paragraph's type the same way, and a Word file keeps its indents. The words and line numbers don't change. If a PDF's layout can't be read, the script is read as plain text, as before, and the progress card says so. On a show that already has a finished read, a new upload keeps the layout but only the scenes that changed are read again; to read every scene with its layout, upload the new draft and press Re-read on the Script page.
An answer on the card's own line is answered for you. When the script check is at least 80% sure, and the line it answers with is the very line the card already quotes, that line is enough: the card is answered for you with Undo, even when the line doesn't spell out the item's name or sits in a scene the card's list numbers differently. Two entries are still joined only when one line names both, or when they are one name written two ways, like "NELL" and "NELL (V.O.)"; "YOUNG NELL" is a casting call and stays a question.
One card per item, and a shared part you might own is asked. When the breakdown has already placed an item on a card of its own, it doesn't also ask whether that item is listed twice. When the line a card quotes names both pieces of a split, the split is made for you with Undo. When the script check thinks two departments could each own a part, or isn't at least 80% sure of anything else the breakdown answered for you, it comes back to you as a question. A join the check finds wrong is undone, or asked when someone's work rests on it.
A set keeps its dressing while the same family lives there. When a scene describes a room's dressing, the later scenes in that room keep it on the breakdown, marked "Carried from scene" with the scene that described it. That holds across the years while the same people live there, read in story order, so a flashback to earlier owners doesn't clear the present family's things. It stops when a scene has its own piece of that kind (a new sofa, that year's tree), when the script says a piece was removed, replaced or broken, when the room is dressed again, or when a different family moves in; their predecessors' furniture doesn't come back with them later unless the script says so. The house itself, its walls, windows, doors and fireplace, stays. Hand props, costume, holiday decorations, and anything eaten, drunk or smoked are never carried, and a carried item never becomes a question. "Not in this scene" takes it off, and the next read leaves it off.

2026-10-21

Read history shows every breakdown on the show. The fold under the script's read now lists each time the script was broken down, newest first: who started it and when, which draft it read, and how it ended. A finished one shows its questions raised, answered for you, closed with a reason and waiting for you. One somebody stopped says who. One that didn't start says why. A restart it carried on through is a line under it. The fold also sits apart from the link above it, padded like the card it's in.
The script check reads scene numbers the way your board does. When your opener is filed as scene 1, every later scene is one up from the script's own count. The check now reads the script with the board's numbers, so the line it quotes comes from the scene the card is about. A fact it is at least 80% sure of, with its line in one of the card's scenes, is answered for you with Undo; under that, the card asks you with the script's answer put forward. A join the read made that the check calls wrong is put back. Two entries that only share a name, like a coat with someone's name on it and a plain coat, are joined only when one line names both. And an answer the check settled reads as what was done, not as the question it was.
A breakdown that goes quiet doesn't hold your questions back. While a breakdown checks its questions after saving the board, it now says "Checking the cards" instead of "Done". If a breakdown ever stops moving, after half an hour the Script page shows your questions again and offers Check them now, which checks every open question on the show in one go, without reading the script again. The person who started the breakdown can press it, as well as anyone who approves the script.
A breakdown never stops because a script is too big. Before it starts, the breakdown now checks how much each of its parts has to send and plans the pieces, so a long script, a big cast or a busy tray is read in more pieces instead of stopping partway. If an entry, or one line of the script, is too long to check even on its own, the rest of the breakdown still finishes and reaches the board, and one card lists what it couldn't check so you can look at it yourself.
A long script with a big cast reads to the end. When the cast check or the check of your questions has more to send than fits in one go, it now sends it in parts, so a bigger cast or a tray with many open questions takes a little longer instead of stopping the breakdown.
A real call stays in front of you. The script read no longer answers a question for you when its quoted line doesn't prove the answer. Two list entries it thinks are one thing are joined only when one line names both. An item is added to another scene only when that scene's line names it. Two spellings of one role are joined only when the read's own check agrees. Otherwise you're asked, and anything already changed stays as it is until you keep it or put it back. One item or two, and which department owns something, are always asked as the department's call, even when nobody has worked on the item yet. A question you were already asked, where someone's work rests on what it names, is never closed for you by a later read. And every card the read closes now records why. When the breakdown closes a question without asking it, the tray now says so: open “Closed by the breakdown” under Answered for you to see each one and why it was closed.
The script check asks of every card what you would. Before a card reaches you, the check now reads it the way a line producer does: is this a real call, does the script answer it, is it the same question as another card, or does its quoted line not show what it says? When a line in the script proves the answer, it's answered for you with that line and Undo, whatever kind of card it is. If someone's work rests on it, you're asked, with the script's answer put forward. A repeat joins the earlier card, even one from an earlier read. A card whose line doesn't match is rebuilt from the board, or left out and counted. It reads what was already answered for you too: two things joined wrongly are split again, and an answer it doubts comes back to you as a question. Each card carries what the check said and why.
Reading a script again tidies what the last read left twice. When you read pages the show has read before, a person or an item the last read listed twice, with one copy now on no scene, becomes one entry, with a note under "Answered for you" and Undo. That holds for two wordings of one thing too, like "Brass lantern" and "Brass oil lantern", when the check is sure they're the same. Things that only sound alike, like a 1962 logbook and a 1974 one, or a flare and a flare gun, are never joined for you. If someone has already worked on either, you're asked once instead. Someone the last read left under a stand-in name like "Unnamed", or with a "(V.O.)" on their name, gets their real name back, with a note and Undo.
A shouted line of action is not a person either. When the script read takes a line of action for someone's name, including one shouted with an exclamation mark, it no longer adds an "Unnamed" person to the cast, gives them a costume line, or asks whether they're the same role as someone else.
Reading a new draft joins what it should, and asks each pair once. When you read the script again, the read's own joins of one thing listed twice are now made, instead of turning into questions because the scenes had moved up one. When the check of your questions finds two entries are one thing and nobody's work rests on either, they're joined for you under "Answered for you", with Undo. Where work already rests on one, you're asked once. And a pair you were already asked about on an earlier read isn't asked again on a new card: the open question picks up the new lines.
Every number on a card from a second read is the board's. When you read the script again, the reason under a card now names scenes the way the board does, even scenes the card isn't about. A card never quotes a line from a scene it doesn't list, a line from the opening is shown as scene 1, and a question that names no scene and quotes nothing isn't put in your tray; the read counts it instead.

2026-10-20

Read again uses the board's scene numbers on every card. When the opening is numbered scene 1 and you read the script again, every card from that read now names scenes the way the board does: its question, its scene list and its quotes. Before, a card from the second read named each scene by the script's own number, one below the board's, so one item read one scene apart on two cards about it. A card also quotes only the scenes it names now, and a question about taking an item off a scene its list doesn't name isn't asked at all.
No question is lost or asked twice. When the script read meant to join two entries and couldn't, you're now asked whether they're one item or two, instead of the card quietly disappearing. One thing is asked about once, not once from each side. A line of action is no longer added to the cast as an unnamed person. A question about two things somebody's work already rests on stays open when you read the next draft. And "Same role as …" on a new name the read asks you about now has an Undo.
Viewing as someone stays on their show. If you're viewing a show as one of its crew and switch to a show they're not on, you go back to being yourself, and the side menu says so in one line. If they're on both shows, you stay as them. Nothing done on the new show is ever recorded against them. Going back to yourself from a view opened from HQ now returns you to your own sign-in, not the sign-in page.
Two shows with one name read apart in the production menu. When two of your shows share a title, the production menu adds each show's code beside its name, so you can tell which one you're opening.
You can stop a script's breakdown while it runs. "Stop the breakdown" sits beside the progress line. It asks once, stops before the next step, and leaves the board, the department lists and the tray exactly as they were. While the breakdown is being saved it can't be stopped. During the departments' check, stopping ends the check and keeps the breakdown that was already saved. The person who started the breakdown, or anyone who signs off the breakdown on that show, can stop it. The card then says who stopped it, when, and after how long, and "Break down these pages" is ready again.
Reading a script again doesn't make the cast or the sheets twice. When you read pages the show has read before, a person already on the cast list stays one person, even when the read meets their name in a new way, and keeps one costume line. An item the read now words differently joins the item the last read made, with a note under "Answered for you" and Undo. If someone has already worked on that item, you're asked once whether they're the same thing. A name the read can't place beside an existing person is asked about once instead of added as someone new.
Two conditions of one thing are never quietly made one. When the script shows something in two conditions, like a scar raw in one scene and healed in a later one, or a waistcoat clean then torn, the read no longer files it as one entry without a word. If nothing rests on it yet, you get two entries, each on the scenes that show its condition, with a note under "Answered for you" and Undo. If money or work rests on it, like a makeup build, a booking or another department's part, you get one question that names both conditions and their scenes.
One thing listed twice on one scene becomes one line. When the script read lists the same thing twice on a scene in two wordings, and it's sure they're one, they become one line on the department's list, with a record under "Answered for you" and Undo. When someone's work is already on one of them, you're asked instead, and the question says so. A question waiting in your tray now quotes the script as it is after a revision, and keeps your note.
One costume line per person per scene. When two cast bookings turn out to be the same role and are joined, a scene they were both on now lists that role's costume once, not twice. That holds whether the read joins them or you answer the question yourself, and Undo puts each booking's own costume line back once.
A delivery mark is not a second person. When the script writes a role's name with how the line reaches us in brackets, like "(on radio)", "(on monitor)" or "(V.O.)", the cast list now shows the role once instead of asking whether they're the same. "Answered for you" says what was joined, with Undo. An age in the bracket, a different name, or anything else about the character still comes to you as a question. Undo on a joined role now gives each booking it puts back its own costume line on its own scenes.
A page break doesn't cut a sentence in two. When a sentence in the script runs over a page break, the read now skips the page number and any CONTINUED line in the middle of it. A prop in someone's hand at the bottom of one page still goes with them into the next scene, and a prop they put down across the break stays put.

2026-10-19

Read again joins a name the last read asked you about. If an earlier script read asked whether two ways of writing a name are one person, reading the same pages again now joins them for you, with a note under "Answered for you" and Undo, the same as a first read would. It still asks when someone's work rests on either name, like a booking, a day-out-of-days line or an answer you gave about them. An answer you already gave, like Keep separate, stays as you left it.
An age stays with its costume. Items named for a child at different ages stay separate. A costume named for one age stays on that age's known scenes, instead of following every scene of the person.
Every card names a scene by one number. When the opening before the first scene heading is numbered scene 1, every scene after it moves up one, and a card's question now moves with its quotes and its scene list: a card about the board's scene 28 says 28 everywhere, never 27 in the question and 28 under the quote. The same goes for an item's note about where else it's carried, a scene the read couldn't read on the finish line, and cards already in your tray, which are put right when the update goes live. A script that prints its own scene numbers keeps them: its opening goes ahead of scene 1, and the card about it says that's because the script numbers its own scenes. You can still number the opening 1 from that card.
Put it back puts an item back where it was, after the opening is numbered scene 1. When the script read moved an item to the scene its line comes from, and the opening was numbered scene 1 afterwards, "Put it back" under "Answered for you" now returns the item to the scenes it was on and takes it off the one it was moved to. Before, it used the old scene numbers, so the item landed one scene early and stayed where it had been moved.
Your tray answers have a way back. Undo after combining a group of items restores every entry and its scenes. "Take it off" on a question within a card takes the item off its scenes, with one Undo to put it back. A second script read keeps your open question and its note instead of adding a copy, and does not ask a question you already answered.
Put it back works on owners, joins and moves after the opening is numbered scene 1. When the opening was numbered scene 1 after the script read, "Put it back" under "Answered for you" now works on an item a department was given, two lines joined into one item, and an item moved to another department. Each goes back exactly where it was, on the scenes the board shows now. Before, all three said the work had changed since and did nothing.
A heading with a year in it keeps its place. A scene headed "1987 -- INT. KITCHEN - DAY" or "INT. KITCHEN - 1987" keeps its place, its time of day, and inside or outside on the board. The year is the scene's period, never its number.
A used invite keeps its details private. Anyone other than the signed-in invitee sees a sign-in message, without the person’s name, role, department, inviter or show title.
A set stays dressed, and people keep what they're holding. When the story goes back to a set, Set Dressing's pieces from the earlier scene are on it again, until the script says one was removed, smashed or is gone. When a scene cuts to the next moment and a person carries on, the prop in their hand and the costume they wear go with them, until the script says they put it down, hand it over or take it off; a jump from day to night ends it. These are never questions. On the Script page each carried item says why it's there, like "Carried from scene 1" or "Still with Odile Vance from scene 1". "Not in this scene" takes one off, "Put it back" returns it, and the next read leaves it off.
Publish stays available after a script arrives in Leeway. A show with Leeway alone can still publish its schedule and change its working week when its scenes come from the script.
"Answered for you" settles only what the script shows. The script read puts an item on a scene only when the line it quotes there names the item; a line that doesn't, like a quote from the scene next door, moves nothing, and its card under "Answered for you" says why: "Not put on scene 5: the line quoted there doesn't name it." A department whose only line doesn't name the item gets no part in it, so nothing there says it has one. A character's own prop is never joined with a piece of the set, so it can't spread to every scene that set is in. Two wordings of a makeup or effects build, like a raw scar and a healed scar, are one build or two, and that's your call, so it comes to you as a question. And when the departments' check finds an item on a scene it's already on, it's the same item, with no second record.
One start button on the Script page, and a page count for PDFs. While the check's questions are waiting, Looks right, start the breakdown is the only start on the page; the file card no longer has a second button that would send you back to the questions. The card ends on the file's details, with no empty strip under them. A PDF's page count shows there once you press start, like "PDF · 1.2 MB · 118 pages". The line saying a full script takes about an hour is gone from the page before a read starts, because nothing has earned a time promise yet.
A chosen script looks chosen. Once you pick or drop a script on the Script page, the dashed drop box turns into a solid card for that file: its name, what kind of file it is and how big, and Replace at the right. Break down these pages sits on the card, under the file. Drop a different file on the card to swap it; the card lights up while you hold one over it. Before you pick anything, the dashed box and "Choose a file" are as they were.
No time estimate while the script is being broken down. "Estimated time remaining" came too late in a read to be any use, so it's gone from the Script page in every form. The bar and the stage you're on stay. Every read now keeps a note of how big the script was and how long each part took, failed reads too. The estimate comes back once a read's first five minutes can call its length right four times out of five.
Pressing start on the Script page shows the check first. The moment you press Break down these pages, the page says "Checking the pages before the breakdown starts. Any questions will appear here." There's no progress bar, percentage or "Breaking down the script" until the check is done and the breakdown itself has begun. If the check has questions, they stay on the page until you answer them, and the progress shows after that.
Cast cards offer the choice you'd reach for. A card naming several people now has "Let me pick" beside "Same person" and "Keep separate": you say for each one whether they're the same person, keep their own booking, or aren't a person to book, and one Undo puts it all back. A card about a handful of one-scene parts offers "Group some of them", and says nothing about ages when the script gives none. A role you already have, seen on a phone or a screen, is a playback shoot: it's settled under "Answered for you" with Undo, and the breakdown sheet's screen column carries that role on the scene. Someone on a screen the read can't name is asked as playback, never "not a person". A passing mention in dialogue of someone already on the cast list, like "the captain", isn't asked. A card about sub-headings inside a scene names each one as "a sub-heading partway through scene 47" and offers "Mark as intercut", which leaves it where it is and marks the intercut on that scene and the one before it. "Book as background" and both answers to a join the checks didn't agree on have Undo too, and under that join the tray says "The join stands until you say otherwise." A card never shows a stand-in name like "Unnamed (35)" or a line of action as a person; it names them by their own cue or as someone in their scene. A surname two roles share, a name with a number on one side only ("GUARD" beside "GUARD 1"), or someone's relation ("WREN'S MOTHER" beside "WREN") is never joined into a role: where they speak, the card asks. "MARA'S VOICE" is still Mara. While a card asks whether "HALLOW'S WIFE" is Captain Hallow, the captain stays on the cast list and on every one of his scenes.
Cast cards ask only what the script can't answer. When the script writes one name in several forms, like initials ("M.V." for Mara Vance), a title before the surname, or an off-screen mark, and every check the read ran agrees, it joins them for you under "Answered for you" with Undo. If another role could answer to the same name, it still asks. A join the checks didn't agree on comes to you as a question: "No — keep them apart" puts each name back on its own scenes. A grown-up whose age the script gives once, where she first speaks, is one role at one age, not a "more than one performer?" question. A line of action about someone already in the scene is that person, not a "who is this?". Someone who never speaks can be booked as background: they come off the cast list and go on their scene as one extra. "Keep separate" counts every person on the card, "Same person" counts every scene any of them is in, and when all three checks agree the card says which way it leans.
The tray asks you only what needs a person. Every question is a casting call, a call a department has to make that the script doesn't answer, a "same thing?" that someone's work, booking or price rests on, something you asked about yourself, or a call about the script's own scenes: an opening with no scene heading, two scenes with one number, or a scene the read could not read. Anything else never reaches your tray. If the script read made a question that fits none of those, it stays off the tray and the finish line says how many, like "3 cards the breakdown could not justify were kept off the tray."
Reading your script does not report a changed shoot day when its scenes stay the same. If you typed scenes and sent the schedule before reading the script, the next send preview recognises those same scenes in the breakdown.
Download feedback stays useful. An interrupted schedule download asks you to try again in plain words. The download-started notice clears after five seconds.
Casting calls come to you as questions. When the script read can't be sure about a person, like an adult and a child with one name, one child at several ages, or whether old footage of a real person needs clearance, it asks you instead of filing it under "Answered for you". The person stays on their scenes while you decide. For one child at several ages, it asks whether one performer plays every age: "No" keeps one booking on every scene, "Yes" books each age on the scenes where the script writes it, and neither takes the child off the cast list. Each of these questions shows the line in the script that names the person. A scene page no longer lists a name the cast list books on that scene, like "ISLA (1YR)", as not on the cast list. Names that are plainly one role are still joined for you: "(CONT'D)", "(O.S.)" and "(V.O.)" marks, a curly apostrophe, a slip of the pen. "Hugh Young" and "Hugh" are one man, since "Young" after a name is a family name, not an age. A role the script names by a line like "BOY IN THE SOU'WESTER" keeps that name instead of "Unnamed".
Worn things are Costume's alone, and a one-word item is named from its line. When Props lists a necklace or a cap that Costume has, Props no longer gets a part in it or a card saying so; a worn thing is Costume's. Props still has its part where its own line says what it handles, like a holster rig on a gun belt. When the script read keeps an item filed under one word, like "dress", and the line says whose it is or which one, the item takes that name, so the costume sheet reads "Odile Vance's grey dress". Undo under "Answered for you" puts the old name back. Where the line says nothing more, the word stays as it was.
The script finish line counts the board and tray. People are the roles on your board. “Answered for you” counts all of the read’s visible records, including those that made no change. The line no longer promises Undo on every record.

2026-10-18

"Lock shooting script" sits with the read. On the Script page it is now in the Current read panel, in the same row as "Read the whole script again" and "Break down new pages". Same button for the same people, same sure-check, same wording. Once locked, the panel says when.
On a show that uses only the schedule, a scene link lands on its row. The scenes the script read made are the schedule's scenes too, so a scene named on Script or in chat opens its row on the Scenes tab, and a character opens Cast. Chat no longer answers what's waiting on you there, and the sidebar reads Today, Chat, The Graph, Script, Schedule and Crew. Scenes you typed before the script was read stay on the board: one the script also has becomes the script's scene on the same day, and the rest stay as they were.
The script read fills in who works on each item. When the read finds one thing listed by several departments, it's one item: Props owns the pacemaker, and Hair & Makeup and Special Effects each have their part, with the line from the script that gives them their work. When the departments' check sees a department working on an item another department already has, that becomes its part too, instead of a second item. Something worn, like a coat or a hat, is Costume's; anything else goes to the department that books it first. Who owns it shows under "Answered for you", in words like "Props owns Pacemaker: Props lists it first; Hair & Makeup and Special Effects have parts.", with Undo, which asks you instead. A department's sighting quoted from another scene goes on the scene where those words are. Each part is under "Answered for you" with its line, like "Special Effects has a part in Pacemaker: “It sparks when he presses it.”", and its Undo takes that part off the item. Anything you've already set on an item's "Who works on it" stays as you left it, and Undo leaves a part someone has changed. A gun belt or a bow tie is worn, so it's Costume's; a gun or a book bag is carried.
One item, every department that works on it. On an item's page, "Who works on it" names the department that owns it and each other department with a line saying what it does: Props owns the lantern, Special Effects rigs the flame inside it, VFX adds the glow. The owning department, the director or a script approver adds a department, changes its line or takes it off. The item then shows on each of those departments' Script page and scene page lists and in their workroom with their line and who owns it, and the owner's card lists the others. Combining two items keeps every department's line, and Undo puts them back. A render's cost is still counted once, against the department that owns the item.
Only your own answers hold an item back. The script read asks about an item when you have answered it: a card you answered, an "Answered for you" you kept, "Not in this scene", or "not an item". An answer the read gave itself no longer counts as yours, so it stops nothing it should settle. When the read wants to change, or add back, an item you took off a scene or said is not an item, it asks you first. A show with thousands of answers reads them once per check, so a read stays quick.
A script read asks you only what it can't settle. An item the script names is added with its line instead of asked, unless someone already took it off that scene or answered it. A call about an item is settled for you only when the read is sure, nobody's work or money is on the item, and you can see the line and undo it under “Answered for you”; otherwise it stays your question. A question that should not have needed you is checked against the script once more. A scene whose items couldn't be read is tried again during the read; if it still can't be read, the finish line names it instead of putting a card on your tray.
A script read changes an item itself only when it is sure and nobody's work is on it. When the read is sure an item should come off a scene, move to another department, or be split in two, and nobody has answered, priced or booked it, the change is made and shows under "Answered for you" with its line and Undo. When it's unsure, or someone's work is on the item, it asks you and the item stays where it was. A scene that names an item in any spelling, like “log-book”, “logbooks” or “knives”, keeps it unless you say otherwise, and an item comes off a scene by itself only when nothing there names it and nobody's work is on it. An item added because its line names it shows that line; one whose line doesn't name it is asked. A question about splitting one item into two no longer stops the read before it saves, and the question names both parts.
One prop is one item, whatever the script read called it. If the read writes "brass lanterns" in one scene, "Odile's brass lantern" in another and "brass storm lantern" in a third, Props gets one brass lantern on all of those scenes. When the names don't say the same thing, like a lamp and a lantern, or the read only leans towards one item, it asks you: "Is “Brass lamp” the same thing as “Brass lantern”?" It never joins a camera with a bicycle because they're both vintage. A flare isn't a flare gun and a 1946 boat isn't a 1953 boat, so pairs like those are asked too, and so is the same lantern later in the story once time has passed between the scenes. The other names stay on the item, each with the line it came from. A join made on a judgment shows under "Answered for you" with that line and Undo. Two people's things are never joined, even when they share a surname, and two names the read isn't sure about stay two items. If the read is sure but someone has already worked on the item, it asks you: "Is “Brass hand lantern” the same thing as “Brass lantern”?"
One thing two departments call by different names is one item. If Props lists a pacemaker and Hair & Makeup lists a chest appliance on the same line of the same scene, the read makes it one item: Props owns it, and Hair & Makeup has its part with its own line. Anything worn goes to Costume, like a gun belt or a bow tie; things carried in the hand, like a book bag or a letter, follow the usual department order. The join shows under "Answered for you" with its line and Undo, which gives you the two items back. When the read isn't sure, they stay two. When someone has already worked on either item, it asks you: "Is “Chest appliance” the same thing as “Pacemaker”?" Same thing makes them one item and gives Hair & Makeup its part with its line.
The departments' check puts what it finds on the right scene and asks you only what the page doesn't say. An item the check finds goes on that department's sheet, with its line, on the scene where those words are, even when it was filed under another scene. A line names an item when it says its main word in any form, like "the lantern's glass", "knives", "batteries" or "oilcan". When the line doesn't name it, like "she lights it", the read adds it when it is sure and nobody's work is on it, and shows it under "Answered for you" with the line and Undo; otherwise it asks you. Anything it found that the script doesn't show on any scene is counted on the finish line. There are no more "Which of these stay" sheets from the check, and no "which scene" questions for an item the read couldn't place: it stays where it was filed. An older sheet no longer asks about an item already on the list or one you said isn't an item. Questions and answers the read already gave you stay as they were when the check finds the same item again.
A name spelled two ways in the script is one person. When the script calls Mara "Marra" once, the read books one person and says so under "Answered for you", naming both spellings and the scenes each came from, with Undo. It only does that for a slip of the pen: the odd spelling turns up in one or two scenes, and the usual one is in at least five times as many. It asks you instead when the names could be two people ("Mara" and "Maria", "Mia" and "Mai", "Guard 1" and "Guard 2"), when both spellings are about as common, or when either person was already on your cast list, has a day-out-of-days line, a booking or other saved work.
Pick the script's answers, then save them together. Your choices stay with you when you reload the page. Review picked answers shows what you chose; Save all applies those answers. A choice that couldn't be saved stays picked, with its reason beside it. The questions you haven't answered stay open.
The finish line counts what's on the sheets. After the departments' check, the Script page says "N new items, M scenes added to items you had" instead of counting every card that landed. The item count in "Breakdown done" is counted from the board after the check too. If some of a read couldn't be filed to a scene, the line now tells you what to press.
The double-order and same-person checks read a short scene whole, and every passage they read opens on its scene heading. Pairs that used to be left for you because their lines were a whole short scene now get an answer.
A day's date is never wiped by a label. Type "Tue 13 Oct" or "13 Oct 2026" on a day and it lands on the calendar, and the page says how it read it. Words the calendar can't read as a date, like "Wrap TBC" or a weekday that doesn't match, save as the day's label, and the day stays on its date with a line saying so. A call sheet's date that isn't a real date no longer clears the day's date either.
A prompt with a broken character still makes the drawing. A pasted prompt carrying half of an emoji, or a hidden null character, used to stop the drawing with a server error. The same goes for the words you give a world build and the name of a new asset. Those characters are taken out and the rest runs as typed.
An answer that kept an item as it was filed no longer says "Change made". When the script read answers a question by keeping an item where it was filed, its card now says "Nothing changed" above "Why we asked", the same as every other answer that left the sheet as it was.

2026-10-17

A show that uses only the schedule now gets the script, The Graph, chat and Today too. Read the script and work through its breakdown, see how its people, places and things connect, and ask about any of it in chat. Scenes link to their row on the schedule. Departments, approvals, review, deliveries, renders and the budget stay with Throughline, and chat says so in one line when you ask for one of them.
A show that uses only the schedule gets the same sidebar, account switcher and sign-out as every other show, with fewer rows. Inviting people and giving them a starting point are on Crew for every show, and how to reach you is in Settings. Old People and You links on the strip board go there.
Crew keep the schedule you published until you publish again. Removing a day, dropping a break, adding a day, or changing a date, call or note shows on your board straight away and reaches crew with the next publish. Until then crew still see every day as it went out, removed days included. Someone given only some days sees only those. Day notes are now part of what you publish. Crew's schedule page tells them the date of the publish they are reading. A show that has never published works as before.
An assistant connected with your own sign-in reads only what you can. Ask it for the schedule and it reads the days you can see and would be sent, so someone on one episode gets that episode's days. It can't open an item you can't open. Someone from an outside company reads the days, deliveries and approvals they were given, and nothing else. Ask it to send a turnover and it tells you plainly that it can't. Access keys read exactly what they did before.
A day's call, date, unit, notes and status keep the copy you opened. If someone else saved that day while you had it open, your save is refused and the page shows what's there now. Ticking a scene on that day adds or removes only that scene, so someone else's order on the day stays.
A day can go between two days on the strip board. Add a day after any day, at the very start, or at a scene, taking the scenes after it if you choose. The days after it renumber and move to their next working day, skipping days off. You see every change before it saves, including whose booked dates move. A day already shot, or being shot today, never moves, and the board says where the new day can go instead. The crew see the new numbers when the schedule is sent again.
An SVG stays a picture. Art can still upload one. Anything in the file that could run, or fetch something from elsewhere, is removed on the way in, and the picture opens as a picture. That holds for a vendor's SVG sent back with a turnover too, and for a simulated picture, which still shows its PNG, JPEG, WebP and GIF ingredients and shows an SVG ingredient by name. A file that had to be cleaned says so on the asset page. A file that never finishes is refused, and so is one saved with editing data we can't open safely: save it again as a plain SVG and upload it again. A PDF on the day opens in the viewer.
The strip board stays put while you drag. The board scrolls on its own, and strips stay in their days however you scroll it, including when you hold a strip near the edge. The unscheduled scenes and the day list stay beside the board, and the day you click in the list stays lit. A scene already on a day is not placed there again, and the board says so right where you let go. A copy on another day no longer jumps. The one-line is a small Export menu that opens over the board.
The schedule pages show each person what they can see. Anyone who can open a scene on the script sees its synopsis and props on the strips and both pages. A company given a day sees every scene on that day with its number, heading, cast and pages, and not its synopsis or props. One page reads as one page, and the page says when it was made, like Fri 16 Oct, 10:56 UTC. A company reads its days as you last published them, until you publish again, and the published shooting schedule prints each day's call.
A company granted some of the days sees only the people on those days. The imported report stays with the office. The strip board and the pages name the same people, and a shoot day on the page reads like Thu 1 Oct. Start and finish on the Day Out of Days page are short dates, and the first days of a longer show are not called a run of their own.
The Day Out of Days dates read as dates. The first and last day under a name are the calendar dates, and a long grid continues on the next page with the names again. A day a person can see keeps its own number.
A printed one-liner keeps each scene's shoot day. Upload it on Scenes and the preview shows the days, their dates, and which scenes sit on them. Nothing moves until you apply it. A day already on the board is matched, not opened again. A scene the script does not have is named and left off.
Removing a shoot day no longer fails when something still names it. A day an outside company was given, a day a pipeline's due dates count from, or a day the Activity feed mentions can now be removed. The company's access to it ends and the record of it stays, the pipeline says the day was deleted, and the Activity line keeps its words. An outside company still sees only lines about what it was given.
API: a delivered file's content can come back as a redirect. Once a show's files move to cloud storage, GET /api/v1/media/{id}/content answers 307 with a short-lived signed address instead of the bytes, as the /media links already do. Follow the redirect (most HTTP clients do by default); don't keep the address, it expires in five minutes. If storage can't be reached, the answer is 503 with a sentence saying so.
A location booked outside its dates is said on the board. Hold a strip over a day the location is not available, and the board names the dates. Drop it and the line stays, and a list above the board gives the day, the date and the scene. The strip still moves. A location with no dates is left alone.
Asking for a sign-in link names your show. The sign-in page says which show it will send a link for. If it doesn't know, and you're on more than one show, the email has a link for each.
A show's admins can make someone else an admin, or stop them being one. It's on the person's card on the Crew page, and on HQ's crew list. The last active admin always stays.
A blank cast number leaves the one already given. Clearing the box, or sending nothing, keeps that performer's number. A locked list says "The numbers are locked." above the numbers, and the boxes stay as they are until the list is unlocked.
The Day Out of Days work total counts a start and a finish. A performer's total on the schedule table now matches the crew download: the first day, the days in between, and the last day all count. A hold still does not.
The office can number the cast from the schedule board. A signed-in member of the schedule office sees the list. A number is a whole number from 1, with one letter if it needs one. A number already given, including one that was changed, stays taken.

2026-10-16

The shooting schedule and the Day Out of Days can be saved as PDFs. Each shoot day names its date and the call, then the scenes on it. A department's items are included when the show has a breakdown. The Day Out of Days file is a landscape grid, performers down and days across, with the dates and the key on one line. A published copy does not include the office's imported marks.
Day Out of Days comes off the strip board. Start, work, finish, a one-day start-work-finish, drop, pickup, hold and travel, with each performer's totals, in cast-number order. A cell the imported report disagrees with is marked, and the legend prints with the grid.
The app tells a browser not to frame it or sniff a file. A signed-in save that comes from another site is turned away. Slack, inbound mail, and an outside tool that signs in with a key are left as they were.
The script read says what it answered before showing why. Factual answers keep their proving words and Undo. Already covered offers existing entries with their departments and scenes to choose from.
Each item on a department's check keeps its own choice and Undo. Combining entries names the whole group; Leave it as it is leaves those entries alone. A change you make stays with your saved answers and can reopen its original question.
Answering one item keeps the department's question above the whole sheet, including after Undo.
A picture sent from the show's storage replaces the stand-in on the file the show already has, including a file filed under the old storage name. Confirming a move of one of those files moves it.
A new draft that numbers every scene its own way is one card. Approve it once and every scene takes its new number, swaps included. A scene whose words changed as well shows the change on the same card. An approver can put the old numbers back from the Script page. When the new draft splits a scene or combines two, the same card says so, and the days, items, notes and answers follow the lines. Anything with no line to follow is asked about once. A draft that only renumbers or reorders scenes asks nothing new, and an edit reads only the scenes that changed.
“What you’ve told us about this show” stays compact after a script read. Its answer count and last change stay visible; open it to see the details or press Edit. Your open or closed choice is remembered for you on that show, while questions and refusals stay visible.
A tool proposed for the catalog keeps its source with that show. The review copy sits with the show's other files, so another show proposing a tool of the same name keeps its own copy, and a deploy throws neither away. A name with a slash in it stays unfiled. The workshop still lists a tool from the tool itself.
A finished departments' check stays finished. Script no longer offers to finish it again or shows an old refusal beside it. When the pages have changed since the script was broken down, Script says so in one line and points to Read the whole script again and Break down new pages.
A deployed service ignores the local walk-on person and unsigned mail, and says so when it starts.
A generated still or clip is kept with the show's other files. The picture made on the canvas, and the auto-gen mark on it, are stored the same way as an upload.
Script's questions offer your show's own departments, and your pick is the one used. Departments are named as your show names them, and titles and graphics items can be booked. Choosing who owns an item moves it there, the card doesn't ask again, and Undo on the saved answer puts it back. A department's check card asks which items stay and lists every one.
Revoking an API key also stops the media links it opened. A ticket fetched with that key stops working at its next use instead of lasting out its week. The ticket and its answer look the same as before.
The show's last active admin can't be paused or taken off the show from the crew list. Make someone else an admin first.
A shoot day can be changed or taken off the board. The office sets its date, call and wrap, and still sees the day's notes. Removing a day asks once, then sends its scenes back to unscheduled, and the days after it renumber. The day's pinned file goes with it. Dropping the break under a day joins its scenes onto the day before. A day already shot stays. A save that lost the race shows the day as it is now.
An uploaded picture is kept with the show's other files. A version, a grabbed frame and a smaller copy of a picture are stored that way, so they are not lost when the app moves off this disk. A failed upload gives its version number back, so the next one on that lane is not skipped. A number a later upload already took stays retired. A drawing that cannot be checked is refused, and nothing is saved.
HQ can send a show's admin a way in. The show page lists each admin with their address and a button to email them a sign-in link, or to confirm an address first when they have none.
A show's admin emails go only to confirmed addresses. An admin whose address hasn't been confirmed gets no show email until they sign in with an emailed link, and HQ's show page names them.
A company seat member whose invitation was taken back after a failed email reads as invited again, however many times the seat was spent.
Someone put back on the show while Slack was taking them out of a channel goes straight back into it, or the office is told to add them in Slack.

2026-10-15

A script read always runs to the end; it never stops partway for cost.
Finishing the departments' check keeps its result when it renumbers the opening scene.
Long feedback titles can be saved and shipped. A vendor request keeps its full accepted name when it reaches HQ.
Put back after Update the N now brings back the steps it removed.
Put back after a template update asks you to look again when it would overwrite a later edit. Your later edits stay as they are.
Finish the departments' check is offered when an older stopped read has not reached every department. Saved parts stay saved; the remaining parts use the same checked pages and instructions.
Finish the departments’ check stays with the saved breakdown. Script offers the newest unfinished check while its saved pages and scenes still match, with no time limit. A later whole-script read that fails or is released leaves that offer available. Finish shows the checked and remaining parts; Read the whole script again stays a separate choice. A released or replaced check cannot be started from a stale tab. An allowance stop points to Finish.
Keep time changes visible beside a new location. Leeway's publish preview keeps the count of other days with new times beside its named locations.
Adding a handoff waits for a retirement already under way. It leaves no wait attached to retired work.
An unsent schedule says so. Readers see “The schedule has not been sent yet.” instead of an empty strip board. The production office keeps its working view.
HQ keeps the script-reading allowance compact. Extra guidance opens under How this works, and Save read allowance is a clear button.
HQ shows script-reading costs in dollars and cents. The spent amount stands beside the show’s allowance, with held costs underneath. Positive costs below one cent say “under $0.01”; when a cost cannot be read, the card says so.
Simultaneous changes at either end of a handoff keep every wait attached to work that is still there.
A crew member who reported a bug hears once when it ships, even when two people at HQ mark it shipped at the same moment.
Slack tells the admins when it needs reconnecting. If the access we saved for an old Slack workspace can no longer be read, the show's Activity says so once and its admins get one email, instead of the app quietly trying again forever.
Taking someone off the show finishes in Slack, even after a hiccup. If Slack didn't answer or a save failed partway, the crew list keeps their line and Take off the show tries the channels it didn't reach. If the server restarts, it finishes them on its own.

2026-10-14

Finish a stopped departments’ check on its own. On Script, Finish the departments’ check keeps the existing breakdown and shows how many parts remain. It stays available after a restart or a later stopped re-read. Read the whole script again is a separate choice.
See a changed location before publishing. The preview names its shoot day and new location.
A refused Script answer stays beside the choice you pressed. On a long card or a phone, the sentence comes into view and your selected answer stays ready to retry. Each question, Undo and Accept all show their own refusal, including Undo under Saved answers.
The script’s first look asks about production choices. Each question is about a choice that changes casting, a department's booking or the shoot. Story facts already settled by the pages stay out of the question tray.
Ask again explains when a scene has left the draft. The question stays open and names the choice to leave it empty. Its scene heading appears once, as a link.
Download the one-line beside Publish. Choose PDF for printing or Excel for final checks. The AD can choose Working or Published; crew download Published. Each file names the show and date, with locations, INT/EXT, weekdays, page totals and confirmed cast numbers. Excel keeps numbers sortable.
Ask again keeps an unfinished reply open. If the reply is cut off or could not be read completely, the question stays open and says that nothing was added. A finished read that found nothing still completes the answer.
Saved answers keeps only completed answers. A refused Ask again stays open with its sentence labelled Last attempt, and its recommendation explains the latest read. Deferred questions stay open too. A completed answer has its own Saved answer line, separate from On the sheet now. Questions the read answered for you keep your saved words beside them; completed answers stay in history. Putting one back reopens its question.
Call sheets and exports keep VFX and SFX in the trade’s spelling.
Move scenes on the Schedule page with a mouse, touch or keyboard. Strips settle into their new day and stay there when you reload. Scroll the page to reach every day, including on a phone. Days already shot are grey and refuse moves. The board keeps day summaries, scene details and numbered cast beside the schedule. Working moves reach crew when the schedule is published.

2026-10-13

Ask again says when the scene still could not be read. One click makes one read, keeps the card open and says that nothing was added, with the choices to ask again or leave it empty. A later successful read answers the card normally.
Each answer on a script card can be put back on its own. Undo sits beside your answer to one question. It reopens that question and leaves the card's other answers alone. If your answer closed the card, its Undo is under Saved answers and puts the item back as it was. Answers on one card come back newest first, and an answer is not put back over a later edit. An answer you gave for two cards at once comes back on both. A change you make with Make the change is listed with your answers, and its Undo reopens the original question.
Ask again keeps a scene's edits while it is being read. If someone changes the scene during the read, the card stays open and says that nothing was added. Ask again reads the scene as it is now.
Invite someone and give them their starting point, on a show with Leeway alone. A People tab takes a name, email and job, and a starting point such as 1st AD, with what it lets them do in plain words. They can change the schedule the moment they open the link. People who have already joined can be given one from the same page. You are offered only the starting points you can give, and each person's card says which one they have, or that one was asked for and not given.
A script card shows the department an item belongs to now. After Make the change, “On the sheet now” uses the department’s name on the sheet; Undo shows the previous department again.
Adding a character finds the one you already have. Adding the full name "Ellis Cole", or "Doc" when the script already uses that spelling for Ellis, gives you the existing character instead of a second one. The same goes for a tool adding a character through the API. Longer labels such as "Ellis Cole — dock look" keep all their words, and adding a look or an item keeps it separate from the character. A name two characters share, or one with an open identity question, makes nothing and tells you what to answer before trying again.
Removing someone from a Slack channel no longer undoes a change made at the same moment.
When Undo on a script card can't put something back safely, it says so and tells you to make the change on the sheet.
A show on Leeway alone sees only Schedule in the side menu, and the pages fit a phone. The other sections of Throughline no longer appear there. On a 390px phone the menu scrolls inside itself, so the page no longer slides sideways.
A department change offered on a script question can be made and put back. The item moves on every scene where it already appears; its scene list stays the same.
The Script page says when a read or question check reached the show's allowance. Earlier work stays saved. It asks the production office to have support raise the allowance before trying again.
A script heading answer says why the set stayed on its scene. Its receipt keeps the heading quote and says nothing changed. Other questions on the card remain open.
A failed save keeps message text and phone numbers out of its error record. PostgreSQL errors retain the information needed to diagnose the failure without copying the row's words into logs or saved errors.
A scene you type or upload keeps its story day, unit and extras. On a show with Leeway alone, the one-line schedule's story day, unit and extras count come across with each scene, and the Scenes form has boxes for all three. Extras is how many people the scene needs.
Publish your schedule from the board, on a show with Leeway alone. A Publish panel shows what would change, publishes it, records how you sent it, and takes it back if it must be.
A show on Leeway alone can publish its schedule. Preview what would change, publish it, and mark it as sent by saying how it went out. Scenes you typed in or uploaded are on the published schedule, and changing only those counts as a change.
On a show with Leeway alone, set the days you work. A Working week panel on the board takes the days the show works and the days it takes off. The days you lay in next follow it. Days already laid in stay where they are.
A script card says why an earlier answer needs reviewing. If its earlier answer cannot be put back safely, the last question stays open and asks you to review that answer first.
A scene's own set can be answered from its heading. When the heading names the set, its script question carries the checked heading and an answer you can put back. An era-only heading leaves the question with you. Matching proposed department changes on two scenes share one answer, including leaving the item as it is.
A question two cards ask is asked once. When the script read asks the same thing about the same item for two scenes, such as which department owns an effect in scenes 8 and 9, the question tray shows it once and names both scenes. Your answer goes on each card. If someone answered one of them a moment before you, nothing is saved and the tray says so.
When the script read answers a question for you, the line it quotes is one it could check. A line too short to check, or one that does not name the item, is asked about again rather than used.
On a show with Leeway alone, you can set your own phone and time zone. A new You tab beside Scenes. Your time zone is what keeps messages from reaching you at the wrong hour.
A show on Leeway alone can bring in its one-line schedule. On the Scenes tab, upload the PDF or text file, look over the scenes it found, untick any you don't want, and add the rest. A scene whose number is already on the show is marked and left out. Nothing is saved until you add them. Each scene keeps its one-line description, and you can type one in yourself.
A name on a call sheet or a Day Out of Days finds the character you already have. If the script already calls Ellis Cole "Doc", a call sheet's "Doc" is Ellis Cole, not a new character. A name two characters share waits for you to say whose it is, and "Sheriff Dace" no longer turns into Dace when "Deputy Dace" is on the show or on the same sheet.

2026-10-12

HQ sets one allowance for the whole script read. The first look and detailed breakdown share it, stored answers remain available, and calls already under way can finish before the next call starts.
The chat fixes the breakdown on a show with no script read. A show that typed its scenes in can ask the chat to move, rename, merge or split an item, or add it to or take it off a scene, and Undo puts it back. Chat changes no longer show up among the script cards.
On a series, a chat change reaches every episode. Ask the chat to move, rename, merge or split an item that turns up in several episodes and it changes it in all of them at once, says which episodes, and one Undo puts it all back. To take it off one scene, say which episode's scene, or it asks you.
Undo removes the scene placement made from a script answer. Answers for scenes already listed on the card can be kept without asking for missing script spans. The other answers on the card stay as they were.
Fix the breakdown by asking the chat. Tell it "the cap gun in scene 9 belongs to the armourer, not props", "these two are the same gun" or "take the tinsel off scene 10", and it makes the change and says what it did. Ask for several at once and each gets its own Undo. If a name fits more than one item, or a scene number is in more than one episode, it asks you which. Press Undo beside the reply, or say "undo that", to put it back.
The schedule board's side menu matches the show's. On a show that has Throughline it carries the same icons and, on a series, the Episodes row. On a phone it no longer takes half the screen.
Releasing a job that stopped partway keeps a newer attempt and any saved changes.
The docs read properly on a phone. Pages start at the top, the page list folds behind a Pages button, and the API page no longer runs off the side of the screen.
A department's second look at the script asks about the things it most often misses. Props looks again for gifts, toys and one of a thing per rider; costume for doubles and handlers; special effects, VFX, set decoration and virtual production for theirs.
On a phone, Send fits and Settings and a department's page fit the screen. The chat bar keeps its Send button on screen, the crew's notes on Settings stack under the list of people, and a department's tabs wrap. Opening the side menu moves the keyboard into it, and closing it takes you back to the menu button.
On a phone, the side menu folds away. Pages use the whole screen; tap the menu button at the top left to open it over the page, and it closes when you pick a page, tap outside it or press Escape. Today, Chat and the chat bar along the bottom fit a phone too. Nothing changes on a laptop or desktop.
A scene's cast is listed once. The scene page shows the cast in its Cast section and no longer repeats it as a "casting" group under Breakdown elements.
The sign-in retry screen fits your phone and offers a fresh sign-in.
Keeping the last script answer closes its card. Undo puts that answer back or says why it cannot. A refused answer keeps the line and scenes it proposed so it can be checked.

2026-10-11

On the new hosting platform, work left by a server that stops during a deploy is picked up within a couple of minutes. Renders, script reads, question checks and department sittings no longer wait for the next restart.
Keeping a script card for an item a scene already lists under another spelling no longer adds it twice. "Handcuffs (rusted)" kept beside "Handcuffs" leaves one line, the one with the note, and one item on the board. Two different notes, like "(rusted)" and "(new)", stay two lines, and reading a scene again keeps every one it finds. Two different names for one item still come to you as a question.
A competing update while you sign in or change your number asks you to try again instead of showing an error.
A script card with two questions about one item asks each one in its own words. Each question gives its own reason, even when the item has no short name yet.
A script card asking how many there are offers a shorter answer. "More than one — name the others" replaces the old ten-word button, and still adds a line for each item you name.
Saved script changes show their time correctly across time zones. Older changes still undo to the same board and time.
Answering a card’s questions in a different order leaves the same final status. A card stays accepted when any change stands; putting every change back closes it as rejected. Each remaining question names its own change, and undoing an earlier change explains what to do next.
The activity feed names Slack channels and people the way you know them. A line about someone leaving a channel, or a Slack account being connected, no longer shows Slack's own ID codes. If we can't name a channel, it says so.
When Slack can't confirm someone has left every channel, the office is told which ones to check.
Answering the last question beside a change the script read made now applies your answer and keeps that change’s undo available. Declined changes no longer count as unanswered questions.
Declining one proposed change keeps the other change’s undo on the tray. Undo also refreshes the card’s heading, and changes that depend on an earlier answer tell you which to answer first.
On a show that schedules without Throughline, signing in opens the strip board. Throughline's pages tell that show's people where the schedule is, with a link.
Two people talking on the same script change at once both stay on its record. Before, the later message could wipe the earlier one from the thread, though it had been sent and answered.
Script facts can be answered within a card while its judgment questions stay open. Remaining judgment answers still apply, and answers show their complete supporting quote. Keep preserves Undo; undoing an answer brings back the questions it made redundant. Using an old Undo button after someone else has restored the question no longer declines it.
Accepting one change on a script card keeps its other questions open. If the first attempt to save was refused, you can still answer each change in turn.
On a show that schedules without Throughline, the first AD types the scenes in. The new Scenes tab takes the number, heading, INT/EXT, DAY/NIGHT, set, length in pages and eighths, and cast. Each scene lands in the unscheduled pile on the strip board, ready to drag onto a day.
Co-heads of a department can share an assistant without seeing each other's private memories. While two heads share a department, the shared assistant uses neither person's private notes.
While you drag a scene on the strip board, it tells you what the move does. The line under the scene gives the new page count for each day that changes, and who now works, is held on, or is off which day. Moving a scene within its day changes nothing, and the line says so.
You can now drag a scene on the strip board to another day, to another place in its day, or to and from the unscheduled pile, with a mouse, a finger on a tablet, or the keyboard. Everyone sees the new order. If someone else changed one of those days while you were moving, nothing moves, and you're told which day changed.
Declaring a standing element, or taking a scene off one, says another save is busy instead of showing a server error. Pressing it again saves it once.
Creating a character, or approving a new one from a call sheet, says another save is busy instead of showing a server error. Trying again saves it once. The external API answers 409 with the same sentence.
Approving a Day Out of Days line says another save is busy instead of showing a server error; Trying again saves it once.
Answering a tray card keeps scene changes that were already waiting to be saved. Combining items keeps those changes too.
Script reads no longer stop when a cue has no reason to be left out.
A late counsel answer cannot replace a newer correction or decision after a restart. Its old cost calculation stays out of the revised record too.
Cards show completed changes as a notice, and keep the right question count as you answer.
A partly answered folded card keeps its correct heading and no longer shows a false wording fault.
The script check keeps a short question when a rewrite would make it harder to read.

2026-10-10

Connected apps keep invitation audit records with completed sends. An invitation waits for your address change and refuses the old address.
An invite that someone has already accepted keeps their place when its first mail finishes.
A busy sheet leaves your unsaved work with you.
An account you cannot disconnect leaves your unsaved work alone.
A saved show setting stays saved when the office makes its next change.
Asking to read one scene again leaves other people's approvals available while the read runs.
A reading leaves alone the questions someone is answering and finishes with the rest still to check.
The Script page explains discarded items in plain words. Each saved reason appears beside the item without an internal code.
Every quote on a question card has been checked against your script. A card shows only the lines we found in the script, at the scene they're in. A question that rests on a quote we couldn't find in the script isn't put to you at all.
Cancelling an episode keeps its recorded wrap dates with it, and the next episode starts with its own record.
Episode times display in your local time correctly. A recorded time keeps its UTC offset when the episode board reads it.
Longer script passages have more room for their cast answers. New discarded items use one of the script read's recognised reasons.
The Schedule page's edit buttons show for the people who can change the schedule, whatever their title. Before, they showed for certain job titles only, so a coordinator given schedule rights under another title saw none, and the wrong title saw buttons that did nothing. If you can't change it, the page says to ask the ADs.
The database rehearsal keeps its restore answer on one line while the app logs. A failed second check reports the difference and gives no pause time; configuration messages stay out of the restored app’s settings.
Lay in a run of shoot days at once. On the board, give the first day and how many, and the days follow your show's working week and skip its holidays, numbered on from your last day.
When someone new takes over a department's chair on a show, their assistant starts fresh. It doesn't carry the last head's instructions, tone or memories, private or shared. The head who left keeps their own assistant and everything it knows, wherever they still work. If nobody heads the department for now, the crew's assistant there is blank too.
A breakdown that was stopped before its questions were checked now says when it's finished. After the questions are checked, the show's feed gets "The breakdown is finished: every question has been checked." Before, the feed and the assistant's status kept saying it wasn't finished.
The schedule has Locations. Add a location by its address and check its pin, name the nearest hospital, the location contact and where the permit stands, and say which dates it's free. Each location lists the days that use it. On a day, pick its locations from the list, two or more for a company move, or turn a typed-in location into a proper one in one press. A location with no hospital shows a warning and never stops you.
A card on the question tray never says the same sentence twice, and every part of a question names only scenes its card shows. When a card holds several questions, each asks about the card's own item and gives its own reason; a later one with nothing new to say tells you its lines are the ones above. A change found twice by the script check is asked once, and two questions about one item never read the same: each says which change it means.
The full Schedule page links to the strip board. "Strip board" sits beside Export PDF on shows that have it. The sidebar's Schedule still opens the full page unless we turn the strip board on for your show.
Each card on the question tray asks one plain question and gives its reason once. A change we couldn't make asks the question again and says we didn't make it. A card about an item with no proper name asks about it plainly instead of repeating the finding. When a card holds several questions, each one shows its own reason and its own lines from the script. A department's card of items the second look couldn't place asks which of them should go on the sheet.
When a change we made to your list can't go on the board, the card asks you the question again. It used to keep saying "Keep one entry for Phone" above "Make the change" and "Leave it as it is". Now it asks "Is “Phone” already on the list?", says we didn't make the change, and leaves the answer to you.
A read stopped while checking its questions can finish from the saved work. Check the remaining questions once to finish the read. While that check runs, the page and question tray say so. The activity feed and assistant give crew the fact without an instruction to press a control they cannot use.
Leeway scene strips show confirmed cast numbers beside the names. A role without a confirmed number keeps its name alone. Reading the board never assigns a number.
Leeway now shows scene strips on the live schedule. Read scene headings, cast names and page lengths in day order, with unscheduled scenes below. This first view is read-only.
A script correction recognises the spelling already on the scene sheet. A correction still applies when the scene sheet uses another saved name for the same prop. If the prop has been removed from the sheet, the correction waits for review.
A name answered as “not a person” is no longer removed while it has saved work on it. Files, versions and approvals count, and so does a place on a delivery or a shoot day, a proposed change, a script change waiting for a decision, a message filed to it, or a connection with a note or a lead time. The question stays open and says what is on it and what to do: combine it with the one it really is, or take that work off first. It stays open too when the app checks the questions against the script and answers one itself, and the other questions are still checked. Before, the answer failed with an error, or left the work pointing at nothing.

2026-10-09

An outside company's seat can no longer read or change the show's element rules, read a delivery's approval number, or file place instructions. On the Script page it read each rule with the scene numbers, headings and reasons the office had written. With the right to approve the script it could file, stop or change a rule, and with the right to set a scene's place it could file or stop a place instruction. On the deliveries list it read the number of the approval holding up a delivery it had been given. Asking to take a row off a scene, or put it back, or to send a delivery, answered differently when the scene or delivery existed than when it didn't, which told it which ones were there. Now it reads only what it was given and those answers are the same, even while someone else is saving a change to that scene or delivery. The office and crew read what they always have, except that a held-up delivery now says what it's waiting on, and an approval from another show is never named.
When the script needs something the list doesn't have, we now ask about it plainly. "Does Props need to source a flare gun for Ines (kept in the tackle box)?", with "No flare gun is on the sheet until you decide." underneath. When the script makes plain who on the crew would source it, the question asks them: "Does the armorer need to source…". Things bought by the amount read right too: "No fake blood is on the sheet until you decide." It says once why we're asking, in a sentence or two, and the scenes it names are the scenes its quotes come from. "Add it" files it under the department you pick, and if we couldn't name the item, you type the name. When we've checked the script and found the item already on the sheet under another name, the card says that, "Nothing changed", and why we asked. A quote no longer says "This wording appears more than once" when it doesn't. Every question card now reads the same way down: the question first, then what's on the sheet now, including a card with several questions, where each one has the sheet line under it.
Every question about your script now reaches the question tray. A question whose wording broke our own rules used to be held back and counted in one line instead. Now it always shows, and you can answer it. Where we can, its wording is fixed first.
Two people with the same name on different productions no longer share an assistant. If neither had a phone number or a confirmed email address on their card, the assistant followed the name, so each could read the other's learned habits and rewrite their charter. Now an assistant follows a person to their next production only through a phone number or an email address they've confirmed. Without either, each production keeps its own. Assistants made before this are looked at when the service starts: one that sits on a single production goes back to its owner, but only if that person is still the one who heads the department (if someone else has taken the chair, it is left alone), and one shared between productions is counted first and split only once the numbers have been read. Until it is split, an assistant two productions share shows none of its charter, voice or notes on either.
A read that finished checking just before a restart now finishes from the saved work. It keeps the scenes it read, and an older read offers Check them now only when that check can start.
When a service rejects its key, HQ and support hear about it. Readiness names the show that supplied the key. Replacing or clearing it lets a successful call clear that show's incident, and repeated refusals send support at most one notice an hour.
A change a connected app makes and its record are saved together. If the show was busy saving at the wrong moment, a new access key, a role change, a pause or removal, or an invite could go through and the app was still told "Something went wrong". For a new access key that could leave a working key whose token was never shown. Now both are saved in one go; if the show is too busy, nothing is saved and the app is told another change is being saved and to try again. A refused request is answered in its own words even when the show is busy.
A script check that couldn't save a batch of questions because someone else was saving a change no longer sits on "running" for good. It finishes and says so. The questions in that batch stay questions, the Script page says some weren't checked against the script, and Check them now reads them. HQ's line says how many were left as questions while another change was being saved. If someone is answering one of those questions at that moment, the others in its batch are still checked and saved. If that answer takes longer to save than the check waits, the whole batch is left for Check them now.
Two HQ administrators saving the feedback-reading choice for the first time at the same moment now both have their save go through, and the later one stands.
Starting another server leaves your private first look and Check them now running. It keeps saved plans and shows current check progress when another read is already running.
A connected app's read now answers even while the show is busy saving something else. Asking for the people on a production, one person, or the production itself, at the moment another change was being saved could come back as an error. It now comes back with the answer.
An outside company's seat can no longer drop files on an item it wasn't given, throw away someone's waiting files, or leave a note on a version it can't see. The item page's drop area used to save them anyway, and a seat could discard a crew member's drop before it was filed. Everyone else works as before.
An outside company's seat can no longer rename, edit, share or upload to an item it wasn't given. Being placed in the department that owns an item used to be enough to rename it, edit, archive or delete its versions and ingredients, share it, answer requests to see it, grab frames, save recipes, make or answer for pictures on it, and change its folders. It could also upload to any item on the show, given or not. Now each of those needs the item to have been given to it. Everyone else works as before.
After the opening scene is numbered 1, questions about your script name the scene as it is on the board. Numbering the opener as scene 1 moves every scene after it up by one. The quotes and the sheet moved, but a question, its answers and its reason could still name the old number, so a card asked whether something was in scene 68 while its quote came from scene 69. Now they name the board's scene, on reads made before this change too. A count such as "names 68 lanterns" or "and 2 more" stays as written, as do times, sums and the number after "page". Where a number could be a page or a scene, the reason is said in our own words instead. An item we add from the script goes on every scene its quotes come from, not just one.
Two conversations suggesting the same new scene number now leave one proposal to approve, not two.
An item stays kept by a department until that department says Yes or No again. Marking it already covered, folding or splitting its card, or handing it to another department leaves the earlier answer standing. A different department’s No still names the department keeping it.
Script-reading progress keeps moving when the time left can't be worked out. Before, the stage and its count could stop updating, which looks like a stuck read. Now the stage still moves and the bar keeps its last number.
An outside company's seat no longer sees approvals, held-up deliveries or other scenes' numbers it wasn't given. On a scene's page it read the title and status of approvals it hadn't been given, and a place instruction that named scenes it hadn't been given. On the Script page it read the full list of place instructions, with how many scenes each covers. On a department's workday it read the name of the approval holding up a delivery it had been given. The Graph also listed which characters were booked in a scene, with a link for each, though the scene's page held them back; and a look-alike item could take the place of a seat's own item on a scene's page. And asking to set a place or an episode on a scene it hadn't been given answered differently from a scene that doesn't exist, which told it the scene was there. Now it reads only what it was given, and those two answers are the same. The office and crew read what they always have.
A delivery can only be held up by an approval on its own show. Filing a delivery with another show's approval, or one that doesn't exist, now says the approval wasn't found and saves nothing. It used to be saved and tied to the other show's sign-off. A delivery already saved that way no longer shows the other show's approval anywhere: its name, its status and its number are left out on the workday, the deliveries list and the send answer, and an access entry made by hand for another show's approval no longer lets an outside company's seat read it. It also stays held, whether or not the other show has signed off, until it's made again with one of this show's approvals, and the answer when you press Send is the same either way.
After a restart, the check of your script's questions picks up where it stopped. If Throughline restarted partway through checking a long script's questions, Check them now used to go back to the first one and check every question again. Now it checks only the questions that weren't checked, and a question checked before the restart keeps its check.
Slack export applications and repeat exports take turns. An application uses the workspace after a move finishes. Running an export twice saves each archived message once; if another change holds the show too long, you are told to try again.
Undo tells you which later change to put back first. If two changes from breaking down the script touched the same scene, Undo on the older one keeps the later change safe and names the item to put back first when its card is on your tray. The tray asks you to check unfinished questions only when you can use Check. Crew see who the questions are waiting for, and Accept all counts each change shown in the heading.

2026-10-08

An outside company can no longer move a scene off a shoot day it wasn't given. Asked to, the assistant says "You can't move scene 12 from here. Ask the production office." and changes nothing. The office moves scenes as before.
If the assistant can't finish moving a scene because someone else is changing one of its days, the scene stays where it was. Ask again in a moment and it moves. A scene an earlier move left on two days comes off the old one when you ask for the new one.
Two people asking the assistant to move the same scene to different days at once leave it on one day. Before, it could end up on both. When the assistant has to wait for another change, it now just says to try again in a moment.
If Slack doesn't answer a request to take back access, or asks us to try later, the request stays on file. Activity says when Slack confirms it.
A connected app that asks to remove or pause someone, or to make a new access key, while another change is being saved is now told to try again in a moment, not refused. That no longer counts against its hourly limit, and the app is told so even when the database is held up for a long time.
Saving while a script is being read no longer gets a busy message for no reason.
A breakdown finishes only after its questions have been checked. A stopped check retries twice, keeps the work already saved, and lets you try again if it still cannot finish. If checking finishes just before a restart, returning to the read now finishes its saved record.
An outside company's assistant now names only the people and items that company was given. In a scene it was given, it lists only those people and items, leaves out the synopsis, and doesn't say how many it left out. Its morning message names and counts only its own items and approvals, without the department's paperwork. It can't talk a script change through. The office still sees every name.
Two people removing each other at the same moment can no longer leave a production with no working administrator. This applies to changes made through a connected app. If the second removal has to wait too long, you're told another change is being saved and can ask again.
An earlier failed email no longer cancels a newer company invitation.
Script-reading progress accounts for the time still needed in each stage. The bar keeps its place when estimates change and leaves room for saving the lists.
An outside company's seat is told only the scenes it was given. A seat given one shoot day and one scene used to read every other scene on that day on the schedule: its number, its slugline and how many there were. It also read the numbers of other scenes an item is tied to on a department's workday, the office's paperwork on that department's list, and what shape the items it hadn't been given were in, on a scene it had. Now it reads its own scenes and the items it was given. On a scene it was given it still reads the names on the breakdown, since the scene's text names them, but nothing more about an item it wasn't given: no link, no status, no condition. The office and crew read what they always have.
Keeping an automatic answer preserves what it said. "Already covered", a hand-off, and "Take it off" no longer tell another department that someone kept the item. An automatic answer nobody has kept leaves the question open.
A department's "Keep none" leaves other departments' choices standing. If another department keeps an item, "No, not an element" names that department and leaves the show unchanged. Each answer records its time, and each section of a combined card keeps its answer and note.
Schedule in the sidebar opens the full schedule page again. Day Out of Days, the Gantt, days and publishing are one click away while Leeway's strip board is built.
Your script's questions still reach you when someone else is saving a change at the same moment. The check of the questions used to stop, say they could not be assessed, and hold them back. Now each one it couldn't check waits for you as an ordinary question.
The crew office can see how a booking would read before saving it. On a person's crew card, entering or amending a booking shows the existing working days, the proposed window and the hold readings. The preview waits for a pause in typing and asks you to check incomplete entries. Nothing is saved until you press Save. Overlapping dates are explained in the preview and when saving.
Picture questions follow what the selected model can use. Leaving one out asks again if another big picture takes its place; resizing is offered only for pictures that can reach the run.
A picture that's too big for the model pauses the run. Leave it out, stop, or make a smaller copy. The smaller copy stays with the show. The original stays as it is.
An outside company can answer a picture question only for an item it was given. A run waiting for that answer stays waiting after a restart.
A missing server address now stops the web app from building or starting instead of sending browsers looking for a server on their own computer.
Pressing Activate twice on a budget plan now activates it once. The second press hears "That plan is already the live one"; before, both presses ran. Two different plans activated at the same moment used to be able to leave both live. Now they take turns, and the later one is told another change is being saved if it has to wait too long.
Giving someone access to an episode or a scene twice at the same moment no longer fails. The second grant finds the first and updates it. If another change to that person's access is in the way for too long, you're told so and can press again.
A department's Workflows tab no longer shows changes or open issues on items you can't see. Before, the Changes list and the open issues still named an item you weren't shown, and who had changed it. An outside company's seat could see them too. You see the changes and issues on the items you can see, and the department's own issues as before.
Two people recomputing the economics from actuals at the same moment no longer see an error, and each is told what it really replaced.
A scene edit saved while a new draft is being read in, or while scenes are being renumbered, is kept. The new page and length, or the new scene number, go on beside the edit instead of over it. If a scene is being saved by someone else at that moment, you're told another change is being saved and can press again.
Two memory saves for one assistant at the same moment now take turns. Bringing a travel kit in, or packing a memory into one, while another change to that assistant's memory is being saved says to try again in a moment instead of failing, and never leaves two copies of the same memory.
A second decision on an access request no longer overwrites the first.
Approve all on a Day Out of Days report now tells you about every line it showed. A line someone else signed off a moment before is listed as already decided, with who did it, instead of going missing from the count. The result has a Close button, and a fully approved one clears itself after a few seconds.
Filing the same render into an item's library twice at once leaves one library file. Its description still arrives when the render's description is ready.
On the new hosting platform, approving every Day Out of Days line at once beside someone editing one of those characters no longer shows an error. The approval waits its turn; if it can't, you're told another change is being saved and can press again.
When several files are dropped and they all get the same refusal, it is said once with every file's name. It used to repeat the refusal once per file. A file that stopped partway through uploading says so and to try again, instead of the general “that file didn't upload”.
When a breakdown finishes, it tells you what it found, how long the whole read took, and where to start. How many questions need you, what was changed for you (each with Undo on the tray), what the departments added to their lists, and any item that turns up again later in the story, with a link to its page. If some questions couldn't be checked yet, it says the breakdown isn't finished.
Undo on a change the breakdown made now puts it back. Before, pressing Undo on one of those changes marked the card and left the change on the list.
The Schedule page loads again for shows that use Leeway.
Marking an item "Not in this scene" for one department no longer loses another department's own sighting of it. Say props and costume both see the desk drawer in scene 1, and you mark it "Not in this scene" for props. Costume's sighting still goes on costume's list, or still asks costume. Before, it did neither.
Undoing a booking change offers to tell the crew member when the earlier change was sent to them. The notice carries the restored terms. The card can refresh its bookings, and an earlier success message clears when another change takes its place.
Starting workflows on many items at once is now for the department head, a coordinator or producer in the department, or someone they've named. Anyone can still start one from an item's own page. The button on the No workflow list says so before it is pressed.
A department can merge a starting point into many workflows at once. Tick the workflows on its Workflows tab, pick a starting point and press Merge in: one grid shows each workflow and the steps it would get, and which already have them all. One press adds the missing steps and the waits that run from them, and Put it all back takes those steps out again. A workflow where somebody has since worked on the new steps stays as it is, and the screen says which one. A merge made from an item's own page can be put back from that page, too.
A department can add another workflow to many items at once. Tick the workflows on its Workflows tab, pick a starting point and review: one grid shows every item, which will get another workflow named for the starting point, and which already have one by that name. One press adds them, each item's first workflow is left alone, and Put it all back takes the new ones away again. An item's own page can now put back a workflow it added, too.
The script check keeps the answers it found when it numbers the opening scene. Corrections follow entries the scene sheets have already combined.
A full-length script that was being looked at when the server restarted now carries on after the restart. The pages you uploaded are read without uploading them again. Before, the look stopped and said the pages couldn't be read in full.
Your browser can now fill in your saved address on the contact step and in Settings → You. A show's location and a scene's places still never take it.
Taking someone off the show now tells you when Slack didn't finish. They are off the show either way. If Slack couldn't take them out of its channels, or didn't answer, the crew list shows one notice that names everyone it didn't finish with, and it stays until you close it. When Slack still has someone in a channel, it names the channel and says to remove them in Slack. If their Slack account is still on, it says where to try again: open Past crew, tick them and press Take off the show. The admins' email says so too, and so does the answer to an app that took them off.

2026-10-07

Two people asking the assistant to add a place or move a scene onto the same day keep both, and two asking it to make the same piece get one.
Two people answering imported day-out-of-days rows keep one booking and its approved days. If someone has already answered the card, the tray says so. If another save is holding the show or booking, the tray says your decision didn't save and asks you to try approving or dismissing the change again.
Hosted services keep sign-in links out of their logs. After a platform move, a resumed read recognises an update, and a blank name from the old host no longer hides the new service's restore points.
Question checking asks you to try again when another change is still saving. Both the Script page and HQ give the same plain answer.
The crew office can fill a booking from a deal memo it is allowed to read. The memo proposes terms and dates; they can be changed or cancelled before saving. Clearing the memo keeps the entered booking details.
The crew office can undo the last amendment or close on a booking. The crew card keeps who changed it and when, and checks that the earlier terms still fit before putting them back.
An outside company's seat can no longer reach an item, a version or a change it wasn't given, through the assistant or the Graph. Its assistant counts and lists only its own approvals and deliveries, ignores a hidden item named on its screen or in its words, and won't start a change analysis; the office does that.
Costume combines undo without moving unrelated links, and scene costume names follow corrected character names.
Confirming a call sheet from Slack and WhatsApp at the same moment saves one confirmation and tells both it worked.
Script-reading progress keeps its place through costume and department-list updates. The remaining-time estimate waits for a known pace, and the second look counts completed windows accurately.
On the new hosting platform, approving a DOOD while a character's costume is being added no longer risks an error when the show has no Costume department yet. Both changes keep their work and share the new department. Nothing changes on today's pilot.
A schedule change is confirmed as it was shown. If someone moves a shooting day after you preview the change, the confirm stops, nothing moves, and the new plan appears in its place to confirm again.
An outside company's seat is no longer handed the assistant's brief of the whole show. Its assistant is told the days and scenes that company was given, and the approvals and deliveries it was given, and nothing else. It no longer knows the story or the plan as a whole. The office is handed the brief as before.
A mistaken booking can be marked entered in error and put back from the crew card. Its dates and terms stay on file, with a record of who changed it. Putting it back checks that another booking has not taken those days.
Undoing a character combination in Housekeeping brings back both characters and their costumes.
On the new hosting platform, a tray answer saved while someone takes a row off that scene's sheet, or puts one back, now stays. If the scene is busy, the sheet says "Another change is being saved. Try again in a moment." Nothing changes on today's pilot.
A bulk change on a department's Workflows tab can always be put back. Change a date so the workflows drop out of your filter and the change's Put it back stays on screen, and on the Changes list too. A change that ran in batches goes back as one, and the head of the department can do it too. If you stop a long change halfway and start again on the rest, you aren't asked the same questions twice.
Putting back workflows you just started leaves no mark on the Sign-off overrides list. A workflow nobody touched goes back as if it had never started, and its history still says it was put back and why.
A bulk change on a department's Workflows tab can cover more than 50 workflows. It's still one review. It runs 50 at a time and shows how far it's got, and you can stop it after any batch and carry on later. Any workflow that changed since you reviewed it is skipped and named, and one step puts the whole change back.
Workflows started from the No workflow list can be put back in one step. Each goes back whole. Any that someone has changed since it started stays, named with a link, and a workflow with a sign-off step asks why first. The Changes list can put them back too.
Starting a workflow on many items at once is now one review. Tick items under No workflow on a department's Workflows tab and pick a starting point: one grid shows every item, which will start, which don't fit, which already have one, and one press starts the ones you leave ticked.
You can set a step's due date, give it an owner, or pin its date from the chat. "Move Barn Interior's Concept to 16 October", "Give Barn Interior's Concept to Ana", "Pin Barn Interior's Concept". The card shows the change the way the Workflows tab's review grid does, and nothing changes until you say go or press Apply. Once applied, the card can put it back.
When the workflows you tick come from different templates, the review grid asks which step stands for the one you picked. "Halo's 'Review' → which step in the Plate shot workflow?", with that template's own steps and "Leave these alone". Nothing changes in those workflows until you answer.
"Select all filtered" on a department's Workflows tab now means every workflow the filter matches. The review grid says so ("Everything matching Late · this department — 3 items · 4 workflows") and changes exactly the workflows it showed you.
If somebody changes a workflow while you're reviewing a bulk change, your choices are kept. Nothing is applied. Every tick and untick comes back where the row still says what it said, and each row that moved comes back unticked, showing what it said when you chose, what you chose and what it says now.
A department can shift dates across many workflows by working days. Each workflow counts on its own calendar, a step's start moves with its due, and the steps feeding it move too. Every change shows before it happens, and one step puts it all back.
"Forget me" now also removes a preference saved at the same moment, and the opener's answer and the walk-on's clear take their turns with every other save. Nothing changes on today's pilot.
On the new hosting platform, changing one person's details (an email, a Slack link or a consent) while they sign in no longer risks an error. The changes wait their turn. Nothing changes on today's pilot.
On the new hosting platform, two people changing the schedule at the same moment no longer see an error, and two days never get the same number. Approving a call sheet, adding a day and making a day today take turns. Nothing changes on today's pilot.
Combining two assets while someone is adding a file to one of them no longer loses the file if the combine is undone. The combine waits for the file to finish landing, so the undo puts it back where it was. If another change holds up the combine for more than a few seconds, you're told another change is being saved and can press again.
The assistant no longer tells an outside company's seat about days it hasn't been given. Not through the show's recent activity, and not through which scenes shoot this week. Each person is told the days they can see; the office is told what it always was.
Two office people re-costing the same card at the same moment no longer get a server error. The second re-cost waits its turn and lands as the next version; if the first takes too long, the second is told "Another change is being saved" and can press again.
Two scout drives of the same route keep their own pictures, so each drive renders the view it was aimed at.
A file the storage agent reports twice becomes one library file, and an image sequence keeps every frame that arrives while it is being added.
Two people choosing what a task belongs to at the same moment leave it belonging to one thing, the later choice. It no longer ends up belonging to nothing.
Approve all tells you which lines it couldn't approve and why, even when another AD got there first.
Sending a scene to a vendor again no longer changes the paperwork an earlier turnover hands over. Each turnover keeps the lineup and cover note that were read before it went.

2026-10-06

You can put a show into wrap and record its wrap dates. Picture, company and office wrap each keep their date, who recorded it and their note. Corrected dates keep the earlier entry.
The script-read card keeps both parts of the read in view, keeps its place when you reopen the page, and leaves room for the second look and card check to finish.
Every character has a costume on the Costume list and in their scenes, even when the script says nothing about clothes. Characters confirmed on review cards or added by import get one too. Folding characters or finding a scripted costume removes an empty automatic duplicate; crew work stays. Renames update the scene lists.
A department can change dates, need-bys, owners and pins across many workflows at once. Tick them on the Workflows tab, review every change before it happens, untick what should stay, and put the lot back in one step.
When a first read stops because the service is busy, the Script page now says so. It shows the same words as the activity feed: try again in a few minutes, and the pages already read are kept.
A long list of scenes on a review card now wraps inside the card and folds after eight. "+20 more scenes" opens the rest, and each scene is still a link.
A finished script breakdown says when its question check has stopped, and crew see only the action they can take.
The question tray refreshes while its questions are being checked and stops when the check stops.
During a read, questions wait until they've been checked against the script. The tray says so in one line, and the questions appear as the check reaches them.
Every question in the review tray now shows what's on the sheet first, asks the doubt plainly, and says what each answer does to the sheet. No line numbers, and a recommended answer says why. A question that can't say those things waits, and the tray and the read's own summary count how many are waiting.
If a read's check stopped or missed some questions, Check them now checks just those. It's on the Script page for anyone who can sign off the script, and the questions show as each one is checked.
A script read's last look puts what it finds straight on the department lists. An item whose line is in the script, in that scene, goes on the list with no question. What it can't place comes as one card per department, to tick the ones to keep. Anything you take off a scene stays off when the same pages are read again, and items a new read doesn't repeat stay on the card until someone answers. The read ends by saying how many items it added and how many questions are left.
A person picked from the list in a demo show can no longer answer or resume a tool's build. Answering starts paid work, so they sign in first, the same as for a change to an existing tool.
When a tool has gone, or changed while a page was open, the page now says what happened. It asks you to reload and look for the tool again, where it used to say "Skill not found".
After you change your email, old sign-in links stop working and a sign-in already using the old address cannot reopen your session. Your previously linked SSO identity still works.
Reopening the Script page during a re-read keeps the existing scenes and exports visible while the new pages are checked, including episode reads and replaced colours.
Reopening the Script page during its first read keeps the progress visible and waits for the checks before showing questions to review.
If someone else is still setting up HQ, the setup screen now says so.
Answering a tool’s questions requires signing in to its show. Answers sent twice resume the build once, and changes stay with the tool’s author or a show administrator.
On the new hosting platform, answering a question in the tray no longer undoes a shoot location someone saved at the same moment.
On the new hosting platform, when two people save the same scene at once, the second waits its turn and is told "Another change is being saved" if it takes too long, instead of seeing an error. Nothing changes on today's pilot.
An optional check no longer stays blocked when its setup stops. If saving the stop fails, another read can try it after the waiting period.
Starting another server leaves ongoing renders, page reads, question checks and department sittings alone. At the next server start, work whose server has stopped is recovered or marked interrupted.
Signing in while changing your email no longer verifies the new address with a link sent to the old one.
Two people putting the same workflow change back move its dates once. Overlapping template merges ask the second person to review again.
When you ship a report, HQ now says how many reporters could not be told. Someone with no number we can text is counted, not silently skipped.
A stopped double-order check can be tried again after its waiting period, even if it couldn't save that it had stopped. The Script page keeps the read's original explanation.
While a script read takes its last look department by department, the rest of the show keeps working. The progress bar keeps moving through that step instead of stopping at 99, never moves backwards, and the line under it says how many departments are done. A re-read, which skips that step, no longer leaves room for it on the bar.
Two people confirming the same schedule change at once no longer shift the days twice. The second is told it was already confirmed, and a confirm that the schedule moved under is refused with a note to look again.
Two people confirming what a new tool reads at the same moment no longer switch it on twice. The second is told it was already confirmed, and a yes given to a tool that has changed since it was waiting is turned away with a note to ask for a change.

2026-10-05

A new show starts with its script drop, shows questions as they arrive, and starts the saved pages when you confirm your answers, including after a reload.
Two re-reads share one double-order check. While one makes the check, the other uses the saved plan, and the Script page says so.
Someone with no number on file no longer holds anything up. They no longer show up as a failed WhatsApp brief (once a number is added, they get that day's brief), and sign-off notices and "your report shipped" notes skip them and carry on to the rest of the crew.
When the double-order check stops partway, the Script page now says part of it was made and it will finish on the next read.
If two people set up HQ at the same moment, only the first becomes its administrator; the other is sent to sign in.
An email change link works once. Opening it twice at the same moment moves the address once. A link for a change that was just called off doesn't move it at all.
Script questions can be checked against the script on the new hosting platform.
Slack changes on a show happen one at a time. Two people connecting Slack, or saying yes and no to the same move, no longer leave the show half-moved. A second install into a different workspace asks before it moves the show.
Inviting the same person to HQ twice at once now sends one working link.
The "busy" notice on the Script page no longer names a company.
A re-read finishes even when its double-order check could not be made this time.
An incomplete cost report keeps its amount held until it can be checked.
The re-read preview keeps scene changes visible when another check still needs to run.

2026-10-04

Cast checks cost about 31% less, at the same or better quality.
The double-order check keeps to your read’s allowance when it has to try again.
The HQ scoreboard explains how its same-item scores count wording differences between two good breakdowns.
The re-read preview now tells you when another check is still needed, and combined questions keep every item and scene they refer to.
The re-read preview checks the same page limits as the read before promising to use saved answers.
Long scripts are read in full, in sections, with questions kept together across the whole story.
A render that doesn't finish says what to do next. Run it again, or pick another model. The message doesn't name the tool.
Saving a scene’s shooting location keeps the sheet changes another person just made. A busy save says what happened and when to try again.

2026-10-03

Script reads, assistant replies and picture-description notices no longer show what our reads cost or their limits, including saved notices and read errors. Scene and retry counts stay visible.
Some models take a picture under 3.9 MB and others under 8 MB; the message names the right one.
Changes to different sheet rows no longer overwrite each other. If another save holds up a change, the page tells you to try again.
Rows taken off a scene keep their department, original name and the show’s date. Scene references open the scene, and a refused change explains who can act and what to do next.
A clip refusal now names the next step. Ask the office to make the clip available in our storage. If storage isn't connected, ask the service operator.
A picture the canvas can't use says what to pick. It has to be a JPG, PNG or WebP under 3.9 MB. Pick another one.
A row you take off one scene stays off the next time those pages are read. Put it back and the row returns. The sheet says who took it off, and when.
Making a canvas preview no longer holds up edits to the same asset.
PostgreSQL restore points say what needs attention. HQ names a missing restore-point folder on PostgreSQL deploys, and failed uploads wait between attempts while the log names the cause and next step.
HQ can see what was bought before a script read starts. First looks and note readings keep their receipts, then join the read they preceded. Purchases still waiting for a read stay visible in Usage.

2026-10-02

Standing dressing, and a prop already on two scenes of a set, shows up on the other scenes of that set in the same era. A prop seen once, a costume, and hair and make-up stay where they were filed. Nothing asks about the carry. A person takes one off a scene with "Not in this scene."
Keeping a scene off, or putting it back, holds when that check drops the scene again. The check does not ask about that scene again. A fresh check that stays quiet puts the scene back on and does not ask. A scene the answer never named is still asked. A different script asks again.
A script check that hits a database problem gives you a plain explanation and keeps the technical detail out of the page.
The activity feed tells a colleague why a script read is held, in the same words as the Script page. The person who started that read still sees the note written to them. A vendor key, a connected app, and a status asked in chat see the colleague's line.
The Script page keeps a held read visible and tells you who can answer. Colleagues and HQ see whose access or script approval needs restoring, with the show named.
A reference clip for a canvas video stays in our storage, the R2 bucket. It is no longer copied to a public file host. If it can't be sent from here, the canvas says so.

2026-10-01

A second upload for an episode being read tells you to try again after that read finishes.
Throughline has a privacy policy and terms of service. Anyone can read them at /privacy and /terms without signing in, and the sign-in page links to both.
Questions from a re-read stay in reach when you close the sheet. The Script page says the breakdown hasn't started and takes you back to the questions in one click, even when the show has no note about how to read its pages.
Two people saving at once keep the work that finished first. Version uploads keep their numbers and references, a delivery is recorded as sent once, and the schedule keeps one current shoot day. Closing a booking checks its latest end date. Repeated crew edits and leaving another person's view no longer add the same change twice.
The read's element check runs on a stronger model: the same answers, sooner, with a few more questions for you to settle.

2026-09-30

A scene taken out of the script no longer shows on that day's strip. Neither does anything that isn't a scene. The same is true on the other places a day is read.
A night-before note stays on the schedule that was sent. Move a day after sending, and the note still fires for the date that went out, at the same hour. A show that has not sent a schedule yet still follows the board, including a date written out in words. Sending does not stop that note when the day was only typed. The Days view stays the board being edited. Search, a scene, a bundle, the routine picker, and a company reading the schedule from outside stay on the days that went out. The assistant still reads the board, and says so at the end of every answer.
A scene marked continuous keeps the time of day of the scene before it. On one script that is scene order. On a show shot in episodes it is the episode's shoot order, then the scene number. A scene filed to no episode comes after the episodes. A night scene followed by a continuous one stays night, including on the board when those two scenes are shot on different days. Publishing the schedule, asked by someone from an outside company, says it is for the ADs and production office. Every workflow door tells that person they reach the work their company was given, not this show's workflows, and to ask the production office.
Sending a review session twice sends it once. Once the session is finished, another outcome stays unsaved and the message tells you to start a new session.
Dragging a shoot day no longer changes which days the money is counting, or those days' dates. Wrapping a day that was on the schedule still moves how far through the shoot you are. The budget and a cost estimate say when that schedule went out. A cost estimate keeps the schedule it was costed against. A show that has not sent one yet still reads the board, and says so. A readiness nudge picks a day by its calendar date.
A second-unit day can have its own folder. Name the unit on the schedule board or on the day's page. Once that plan is sent, a new folder for the day carries the unit and the date that was sent. Moving the day and not sending it does not change that name. A day with no unit is still main unit. A name that is not letters, digits and spaces is refused. A folder that already exists keeps its name, so a show already shooting second unit keeps the names it has until someone moves them by hand.
A short scene quoted by several entries is still checked. The check for the same thing listed twice, and the check for one person named two ways, used to split a short scene across many requests. The first of those checks could stop at its request limit with most of the pairs unasked. They now shorten those passages so more of the scene fits in a request, and they still never send the whole scene. When shortening isn't enough, they ask the entries in smaller groups, up to that limit. A comparison that is the whole scene on its own still stays for you.

2026-09-29

Add a dates sheet to the chat. Pick it with the Sheet button beside Send and say what it's for: "update my dates on turnover one to match this". It's checked the same way as on a Workflows tab, only against the items on that date or in that department, and the reply takes you to the column check. If the chat can't tell which items you mean, it asks, with a button for each. Nothing goes in until you approve it.
Ask the chat about a show's workflows, and change a due date or add a wait from it. "What's late in Construction?", "what is Barn Interior waiting on?", "why is its date the 14th?" and "what does Blueprints feed?" answer from what you can see, with links. "Barn Interior is needed by 4 October" or "Fabrication has to wait on Design sign-off" shows the change first, with how many dates move. Nothing changes until you press Apply or say go. With two changes waiting, go asks which one. These questions and changes never sign a step, skip a sign-off or take a wait away. For those, the chat tells you where they're done.
Today, the approval list, search, the activity feed, and the show summary no longer show an outside company what it was not given. The script list and a pending script change now stay with what that company was given. Search no longer turns up a delivery or an approval they weren't given. Today's count of pending script changes counts only the ones on scenes they were given. A note that isn't about something they were given stays off their feed, including a notice about the whole show, and so does a line about a scene they weren't given. Today uses the same counts. A scene they were given still shows, including the names of what's in it. An approval in their own department stays off the list unless that approval was given to them. Today's list of what needs them leaves off an approval they were not given, including one in their own department, and signing that approval is refused. Crew and the office see the same lists they see now. A company key still reads the whole script.
Every Slack card says a refusal on Slack's page never comes back. Beside the address this deploy sends, the card always shows the same line: if Slack says it doesn't know this address, add it to the Slack app's redirect URLs, then connect again, because that refusal happens on Slack's page and never comes back here. The card cannot tell whether Slack has already refused. HQ readiness says the same sentence.
A seat notice no longer quotes the mail service. When mail doesn't go out, the notice says the address was refused as inactive, that the service couldn't be reached, or that it wouldn't take the message. It no longer includes the service's status code or its own sentence. That is true for someone already on the show, and for a new person whose link stays on screen. The service's own words stay in the log. HQ is still told to spend the seat again.
Nothing is added unless somebody on the show is named. A WhatsApp number the show doesn't carry is still turned away. If a number leaves the roster after the message arrives, nothing is added. HQ staff who aren't looking through a crew member are told to open the show as that person before they add. Looking through a crew member, they can. Someone who isn't signed in still gets the sign-in sentence. A company write key is unchanged.
The Workflows tab lists your spreadsheets without reading every sheet on the show. Forty of yours used to take forty-six reads. One and forty now take the same seven. It asks for yours. Somebody else's sheet never appears on yours.
A long list of approvals in Slack still arrives. When a lot is waiting, the reply used to fail and nothing showed up. Now it shows the first ones, and a line saying how many more are in Throughline, with a link to the show.
Approving in Slack, or sending one back, no longer shows an error when Slack is slow. The decision has already landed. The note that it went through follows, so a slow reply doesn't look like the decision failed.
Who may add a shot is who the office gave that permission, not the job title. A production coordinator whose permission was taken away can no longer add a shot. A grip the office gave it to can. Somebody whose credit also says production coordinator, without that permission, cannot.
The show's Slack card says what a missing permission costs. When Slack is connected but a permission is still missing, the card used to print the permission's raw id. It now says what that keeps the show from doing, one sentence each, and names Reconnect. One this version doesn't know gets one plain sentence instead of the id.
The Slack card says what to do when the connection can't be finished. Beside the address this deploy sends Slack, the card always says: if Slack says it doesn't know this address, add it to the Slack app's redirect URLs, then connect again. It no longer shows an error code, or a path to a setup file.
A channel @-mention is answered where you asked it. Ask the Slack assistant in a new message and its answer now shows up right there, for you alone. Until now that answer never appeared. Asking inside a thread still answers in that thread.
A breakdown lists each piece of work on its own row. Two things to build, buy or shoot used to share a line joined with "and", "&" or "/". Now each gets its own row. One thing called two names, a rig with its parts, a clip's playback with its license, or a set bought and handled together still stays on one row.
HQ's scoreboard shows why an element was missed. Five new rows: the items in the answer key the script never names, the ones the breakdown found something like but didn't count, the ones the script names and the breakdown missed, the ones with no script text to check, and the recall over the items the script does name. The first four sit under Coverage context, so a fall in one never shows as a warning.
The checks for the same thing listed twice and for one person named two ways now read the script around each line they cite. Each used to send one cut quote per cited line, and for a name that was often the name alone. Now each sends a few whole lines around the line, never a whole scene, and a character's name always comes with the speech after it, past any parenthetical, one line or wrapped over several, or with the parenthetical alone when that's all there is. Each check's sign-off now says exactly what it sends, and a sign-off given to the old words doesn't carry over to the new ones.

2026-09-28

Drop editorial's turnover EDL on a department's Workflows tab and check the shot list before anything is filed. It asks where editorial put the shot IDs (their markers or comments) and the frame rate, then lists the shots: already on the show, new, and ones it couldn't match. Shot names are always editorial's own. A cut without a shot ID never becomes a shot. It's listed under "Not VFX shots", or under "Already on the show" when its clip name is a shot you have. File shots one at a time or everything that's ready, into the department that took the EDL, and put them back if you change your mind. Drop-frame EDLs aren't read yet, and shots aren't yet matched by reel and timecode. The file itself isn't kept.
Drop a VFX dates spreadsheet on a department's Workflows tab and check it before anything changes. It reads the first sheet of an .xlsx or a CSV and asks you to confirm which column is which. It remembers your answer for the next sheet with the same headings, but you still confirm it. The rows then come back in three groups: matched, best guesses to check, and ones it couldn't match. Fix a date, or pick the right shot or step for a row, approve rows one at a time, or put in everything that's ready at once. A best guess only goes in once you've ticked it. A blank cell leaves a date alone, and a dash clears it. If someone changed a date after the sheet was read, it isn't overwritten; the row waits for you to look again. Only the person who dropped a sheet sees it. Put the dates back with one button if you change your mind. The file itself isn't kept.
The department page, a routine, and a refused signature no longer name an approval an outside company was not given. On the department's page, an approval in that department used to show by title, with the scene, and a delivery the company was not given showed in what needs attention and in the turnovers, including one coming in from another department. A routine that reads approvals waiting did the same. Refusing a signature on a decision named the approval and who chairs it, and a decision the seat itself chairs but was not given said the call was the seat's. An approval the company was not given that was already decided, or that moved while you were reading it, used to say so. Those stay off the page and off the routine now. An approval the company was not given answers as one that is not on the show, including one already decided and one that moved. Someone who was given it still hears that it is already decided, or that it moved. The department head still sees the page, and someone on the show who is refused on a decision they don't chair still hears whose call it is.
A seat HQ sets up for somebody already on the show goes to their inbox. When HQ seated an outside company's member at an address that was already a crew member's, HQ was shown a link that signed in as that crew member. Now the link is emailed to them and HQ sees "Sent to their inbox." If the show has no email key yet, the seat is turned down and HQ is told to connect one; if the email doesn't go out, no seat is left waiting. A seat for somebody new to the show still shows HQ the link.
Ask the show's Slack assistant in a channel, and send an approval back from Slack. Mention the app in a channel it has been added to — "@Throughline what's waiting on me?" — and it answers in that thread, to you alone, since the answer is worked out from what you can see. An approval card in Slack now offers Send back beside Approve: it asks what needs to change, as the web does, and sends the step back with your note.
A scene that would fill one request is asked across the requests it takes. On one roster, one scene takes three. A comparison that covers a whole scene on its own is left out, and the check asks the rest. A line is counted in the scene it falls in. A request that never goes out is not charged.
An outside company's seat can no longer add a shot or an asset to the show. The workroom button, the assistant, and a mention in Slack are turned down, and the seat is told the office shares what they can see. A shot from the workroom button is for the production office or for VFX. Crew still add an asset, in the department they named. A company write key is unchanged.
When mail can't go out, the reason no longer repeats an address or a link. If the mail service sends back an address, a web link, or the invite link, that text stays off the note you see, and off the refusal when the person is already on the show. That includes an address written with letters outside plain English, and one with no dot after the @. The sentence no longer names the mail company or an internal error, and it says "mail key" throughout.
An outside company's seat with no days is told so when a Forge tool asks for today's schedule, instead of being handed an empty one while the show is shooting.
Asking to change a tool that hasn't been built yet says so in words. It used to name an internal status. Now it says the tool hasn't been built yet, so it can't be changed. A tool waiting for someone to confirm what it reads says that. Pressing Run on a tool that isn't ready says it isn't ready to run yet, without naming an internal status.
Your next day's call shows up even when the office is on a different one. If you're with an outside company and the office is shooting a day you weren't given, a tool or a routine that reads today's call sheet now shows your own next day, however far off. A day already wrapped doesn't count as that next day. If you weren't given any day and the office is shooting, the tool won't hand you an empty call sheet — you're told you can't see today's. A show with no shoot day still comes back with nothing to read, and a routine with no day of your own still says nothing is coming up. The office still sees its own day.
Two shoot days added at the same moment take two day numbers. Adding two days together used to give both the same number. The second one now takes the next number.

2026-09-27

After this upgrade, removing an asset no longer lets a new asset inherit its old references.
A Forge tool is rebuilt one at a time, and two rollbacks at once keep both. Asking to rebuild a tool that was already being rebuilt used to start a second rebuild, and whichever finished last kept the tool. Now the second is turned down in words: "This tool is already being rebuilt or changed. Wait for that to finish, then ask again." The same goes for a rollback, a change, answering the Forge's questions, or the tick on a tool's data while it is being rebuilt, for editing its fields while it is being rebuilt or while a change is being made, and for a rebuild while a change is being made. Two rollbacks asked for at once both land, one after the other, and the history keeps both. Answering the Forge's questions about a new tool that has since been built another way is turned down, instead of building it again over what it became. A rebuild that stops partway lets go of the tool after 30 minutes.
A file link stops opening for someone who hasn't started yet. A person whose first day is still ahead could open a file from a link they already had. Now that link stops, the same as it does when the office pauses someone or takes them off the show. The show's pages already kept them to getting set up until their first day.
Your next day shows up even when the office is on a different one. If you're with an outside company and the office is shooting a day you weren't given, a tool or a routine that reads today's schedule now shows your own next day, however far off, and only the scenes you were given. With no days of your own, it reads as a show with nothing coming up. The office still sees its own day.
If an asset is combined or removed while a library file uploads, the upload tells you what happened and leaves no unsaved copy. When the combined asset is still available, the message names where to upload again.
When departments answer a counsel together, their approvals, corrections and stances stay on the record.
An app an outside company's seat connects reads only what that company's key reads. Somebody from an outside company could connect an app to the show by signing in, and the app then read the whole show: every item, scene and shoot day. Now it reads only the scenes and characters the seat was given, and in them only the items, shoot days, approvals and notes the seat was given too; every other read turns it away and says why. The consent page tells them so before they approve it. Anybody else's app reads what it did before.
An outside company's seat receives only the show settings its screens use. A seat used to be sent the whole show's settings, including the office's budget, paperwork and assistant settings, even though its screens never showed them. Now it gets the look of the app and the show's time zone, and nothing else. Crew and the office get what they got before.
Accepting the office's invite shows you the day being shot. If your address already had a seat waiting with an outside company, accepting the office's invite used to leave Today saying there was no shoot day, even when the show had one. Now you come on as crew, and Today shows that day.
An outside company's seat opens only the files for what it was given. A seat set up for an outside company could open any file on the show from its link: another day's call sheet, pictures of items it wasn't given, another company's turnover paperwork, files in the show's storage folder. Now it opens a day's files only for a day it was given, and an item's pictures and generations only for an item it was given. The production office's paperwork stays with the office. Its list of generations shows only its own items. Take a day or an item away and its files close straight away. After this update, a page left open may show broken pictures until you reload it.
Picture-description limits save reliably when two HQ admins change them together.
Picture descriptions share one daily allowance, however many servers are running.
Two changes to one Forge tool at once no longer lose one. When two people asked for a change to the same tool at the same moment, both were told it was done and one change quietly disappeared. Now the second is turned down in words: "Somebody changed this tool while yours was being made, so your change wasn't saved. Look at the new version, then ask again." A reel's "seen by" also counts every load and every piece opened, even when someone has it open in two tabs.
A counsel keeps the first decision when two people answer together, and a retry runs its deliberation once.
If paperwork has to wait too long to save your decision, the card asks you to try again.
When two people answer the same paperwork change, the first decision stays. The second person sees “This paperwork change was already approved.” when the approval was saved first.
Inviting someone who's already on the show now emails them directly. Before, the office was given a sign-in link to pass on. Now the invite goes straight to that person's inbox. If the email can't be sent, the invite isn't created and you're told why: no mail key, no return address, or the mail service's own reason. A seat that Critical Period set up for an outside company is re-sent by Critical Period, not by the office.
A one-time sign-out the first time your address is confirmed on a show. The first time your email is confirmed on a production — through a sign-in link, an invite you accept, single sign-on, or switching into that show from another one — any sign-in already open on that seat is closed, and you sign in again. It keeps a seat set up in your name from being used to reach your real seat somewhere else.
Only you can change the email address on your card. The production office and a show's admins can no longer change someone else's email address, so nobody can move your mailbox to one they read and sign in as you. For a crew member with no address yet, the office can add the first one from their card in the crew directory: we send a link to that address (or give you the link to send, if the show can't send email yet), and it goes on the card once they open it, with a note saying who added it. Someone who runs part of the show gets theirs from Throughline support. The office sees a change that's waiting on the card and can call it off, and Throughline support can still change an address for you.
An outside company sees only its own days and scenes everywhere, not just on the schedule. Last week the schedule board and Today were narrowed to the days a company was given. Now the rest of the product is too: search lists only their own days and scenes; their department page and the day cards on it show their own day and only the scenes on it they were given; the brief they'd get, the daily brief, and the operations digest name only their own days; a scene's page shows the day it shoots only when it's one of theirs. Ask the assistant "what's the call time on day 12?" or "what are we shooting today?" and it answers from their own days and scenes — a day they weren't given is answered as a day on another show. Crew and the production office see exactly what they saw.
Quiet hours follow the day being shot, the minute its call changes. Pulse sends nothing earlier than an hour before call on the day being shot, or later than an hour after wrap. Move that call and the quiet hours move with it straight away, whether or not you've published the schedule. A change to any other day leaves them alone.
Lead times applied from an item now stay on the template.
Switching a consent off switches it off, even after a double tap. If you switched one of your consent cards (email notifications, WhatsApp, text messages, or learning how you write) on twice at once, with a double tap or from two tabs, it was kept as two yeses, and switching it off took back only one, so the card came back on. Now two taps at once are one yes, switching off takes back every yes, and so does Delete everything Throughline learned about me. If you switch a card on and off at the same moment, the one you pressed last is the one that stands. If this already happened to you, the card is off again from the moment you switched it off; if you had switched it back on since, switch it on once more.

2026-09-26

An outside company sees only its own shoot days on the schedule and Today. A company given some of your shoot days used to see every day on the schedule board, with each day's call, notes and call sheet, and Today showed them the office's day. Now the board shows them only their days, and its header counts the days they were given. Today shows them their next one, or nothing when they have none, and their own days on Today list only the days they were given. Opening a day they weren't given, from a link or its page, gets the same answer as a day on another show. Deleting a shoot day now also removes its files and the outside companies' access to it.

2026-09-25

A script read finishes in about half the time. Our feature-script check took 49 minutes, down from about two hours, with cast and department breakdowns read alongside each other.
Everything filed under a department now shows on the one-liner. An element filed under a department like Construction, Scenic Artists or Music & Scoring, from your note or an instruction, used to be saved on its scenes and left off the one-line schedule. Now any department with something on the sheet gets its own column after the usual ones, under your show's own name for it, in the spreadsheet and the CSV. A car filed under Picture Vehicles shows under Picture Cars. Anything filed under Locations shows too, apart from the scene's own set, which stays in its own column.
An empty file, or one over 2 GB, dropped on an asset's library is turned down on its own. When you drop files or a folder on an asset's library to have them sorted, a file over 2 GB or an empty one now says why it wasn't taken, and the rest of the drop is sorted as before. If something goes wrong partway, none of the drop is left behind, so you can drop it again.
Two library files with the same name keep their own files, and a big clip no longer stalls the server while it's made ready to play. Adding a file to an asset's library with the same name as one already there used to replace the first one, so both showed the second file. Now each keeps its own. A clip added as a version or to the library is now made ready to play in the browser a piece at a time, so a big one no longer takes the server's memory with it. The library now refuses a file over 2 GB and an empty file, as the Versions panel does.
A long script read no longer holds up everyone else's saves. While a long script read saved its breakdown, anything else being saved waited for it: answering a card, saving a note, a Pulse message going out. On a very dense script the wait could run long enough for those saves to fail. Now it's a couple of seconds.
Adding a version no longer holds up everyone else's saves. While a new version's file was being saved on the Versions panel, or sent in through the API, other saves on every show waited, and a large clip could make them fail. Now they go straight through. For integrators: the API's version upload (POST /api/v1/assets/{id}/versions) now refuses a file over 2 GB (413, "file is over 2 GB") and an empty file (422, "file is empty"), with nothing kept; everything else it answers is unchanged. The Versions panel refuses the same two, in its own words. And a version added in a second lane with the same file name as the first lane's no longer replaces the first lane's file.
Adding a world by hand no longer holds up everyone else's saves. While a world file added by hand on the World panel was being copied into place, other saves on every show waited, and a large file could make them fail. Now they go straight through.
A world you add by hand opens in the world viewer. A world file added to an asset by hand used to be refused when the viewer asked for it, even to the show that added it. It now opens for that show, and stays closed to everyone else.
Invite them as usual stands while a held seat is being confirmed. If one coordinator chose Invite them as usual on a held Slack seat while another pressed Confirm & provision, and Slack then failed or turned the seat down, the seat went back on hold for its old reason, as if nobody had chosen. Now the choice stands. The seat comes off the held seats, the activity says who chose Invite them as usual while the seat was being confirmed, and the person confirming is told that somebody chose it and to press Provision now to try again. If the seat was marked sensitive again after that, it stays held for the new reason.
A world you add by hand, and a file dropped on a shoot day, go straight to storage. A world file added to an asset, and a file on a shoot day's page, now upload straight to the show's storage, so a big splat or a recce clip no longer passes through the server on the way. A file on a shoot day opens just as before.

2026-09-24

The read’s progress and time remaining now count the checking stages too.
WhatsApp messages and Slack clarification taps are accepted while the assistant prepares its reply.
A new version or library file goes straight to storage. On an asset page, a new version or a library file now uploads straight to the show's storage, so a large still no longer passes through the server on the way. It opens just as before.
Today says when the schedule you're looking at was published, and how many days have moved since. Under the day on Today you read "As published Tuesday 2:04pm", in your own time, and "2 days have moved since this went out" when the board has changed since. The day shown is always the one being shot, whether or not anyone has published, and it's shown as it went out, unless it was added since or isn't one of your days, when it's shown as the board has it. A schedule published before call times were part of it shows the board's call instead, and Today says so. Before anything is published, Today says so. When you publish a new call time or place over a schedule that has call times, the preview now says how many days have new times or places.

2026-09-23

Publishing the schedule now counts a changed call time, wrap, place or confirmed call sheet as a change. Before, publishing after changing only one of those said nothing had changed. If your last published schedule is from before this change, a change to only those is still refused, and you're told why: the next schedule you publish after a day or a scene changes will include them. And what's published is what the preview showed: if someone changes the schedule while you're reading the preview, nothing is published and you're told it moved.
Every row on the sidebar has an icon, and more buttons do too. Episodes, Crew, Deliveries, Budget, Forge and Routines now have icons like the rest of the sidebar. Sunrise and sunset times show sun icons, and the buttons to archive, restore, delete, upload, retry, undo, flag, remix, flip, add to a reel, zoom, sort and pick favorites are drawn icons now instead of typed characters.
New icons on the sidebar and across the app. Today, Chat, The Graph, Departments, Script, Schedule, Approvals, Review, Jobs and Show setup each have an icon on the sidebar. Across Throughline, the arrows, the close, edit and add buttons, downloads and open-and-close markers are drawn icons now instead of typed characters, in the colour of the words around them, in light and dark.
A restart during the first look at your pages no longer loses them. If Throughline restarts while it's taking its first look at your script or at new pages you've just uploaded, it now carries on by itself, and you don't upload the pages again.
Times on Today's notes card and on the Jobs page's On its own tab are right wherever you are. When notes came back on your version, the card on Today showed the time they arrived wrong for anyone not on UTC: seven hours out in Los Angeles. On Jobs, a new line under On its own said "just now" for seven hours in Los Angeles. Both now show the right time wherever you are.
Matching someone on their crew card during Provision now sends them no second invite. On a managed Slack workspace, if an admin matched a person's Slack account while another coordinator's Provision now was running, the person could get one more Slack invite, and the Coordinator queue showed it as waiting. Now Provision now sees the match and puts them in their channels instead.
Invite them as usual during Provision now leaves no held seat behind. If someone chose Invite them as usual on a sensitive seat while another coordinator's Provision now was running, the Coordinator queue could show that person as held again, for a reason that was no longer on their card, even after the next Provision now had seated them. Now the seat is either invited as usual or taken off the queue, whichever happens first. A hold also shows the reason the seat has now, if it was changed while Provision now ran.
The time on Today's Pulse line is right wherever you are. Where we are, on Today, shows one thing Pulse flagged for you and how long ago. For anyone not on UTC that was wrong: in Los Angeles something flagged this morning still said "just now" seven hours later. It now counts from the right moment wherever you are.
Send a Slack invite again. On a managed Slack workspace, the Coordinator queue on the Messages panel lists everyone who's been invited to Slack and hasn't been matched yet. If someone's invite never arrived, press Send the invite again on their row and they're sent a new one. If Slack doesn't answer in time when an invite goes out, it may still have gone, so it stays on the list with Send the invite again on its row, and Provision now doesn't send another. Before, Provision now treated it as turned down and sent another. When you send an invite again or confirm a held seat, you're told if Slack didn't answer.
The production office can reopen an accepted week. Open the department's Reported figure on the Budget page, choose Reopen this week and give a reason. The figures stay as filed, and the activity records who reopened it and why. The department head can choose Check week on their closed card and file again, keeping any unsaved lines.
A failure on our side reaches your screen with a reference. When something goes wrong in Throughline, the page can now show what happened and the reference to quote when you tell us.
The production office can accept a department's week on the Budget page. A department head could file and submit their week's spend, but nobody could accept it anywhere in Throughline. Now a department with weeks waiting says so beside its Reported figure on the Budget page, "1 week to accept", and each waiting week has Accept this week for the production office — the line producer, the UPM, the production accountant, the director and the admins. Someone who can only see the money can't accept. Accepting asks once, showing the figure you're accepting, because only the production office can reopen it, and it accepts that figure or nothing: if the department changed the week after you opened it, you're told it changed and nothing is accepted. If someone else in the office accepted it a moment before you, you're told it's already accepted. The department head's form for that week then reads accepted and closed. If they still had it open and save it, they're told the week is already accepted, and the form closes on the figure that was accepted.
Someone waiting to accept a Slack invite gets one invite. On a managed Slack workspace, anyone without a Slack account was sent another email invite every time someone pressed Provision now, until an admin matched their account. Now they get one. The held seats on the Messages panel show who's been invited and what to do once they accept, and on the crew contact card the control is now called Match their Slack account.
Invite them as usual takes a seat off the held seats. If you chose Invite them as usual for someone whose Slack seat was held, they stayed on the Messages panel's held seats, and Provision now invited them anyway. Now they leave the held seats in the same step, and the card on the People page says so. If someone presses Confirm & provision on the row afterwards, they're told somebody already chose Invite them as usual, and to press Provision now instead.

2026-09-22

A held Slack seat that Slack won't take stays held. If you pressed Confirm & provision and Slack turned the seat down, the seat quietly left the held seats with nobody invited, and nothing said so. Now it stays on hold and you're told. If it's something on your side, like a crew card with no email address, you're told what to fix. If Slack took some of their channels and not others, you're told which are still missing and to press Provision now to finish. If Slack sent them an invite, you're told to ask an admin to match their Slack account on their crew contact card once they accept, then press Provision now. If someone chose Invite them as usual before you confirmed, the seat doesn't go back on hold, and you're told to press Provision now to try again.
Cast review questions use the current script. An earlier script no longer adds a question about names already agreed to be one person, or causes a question to be skipped.
Viewing as someone can't set up or take down storage, move the show's Slack, or share a version. This follows the change to keys and Slack accounts earlier today. While viewing as someone you could still move the show's Slack to another workspace, set up storage for the show or take it down, and share a version or a reel, or send a sitting, in their name. Moving Slack and setting up storage lasted after the view ended, taking storage down can't be undone, and a share can't be taken back. Now they're done from your own sign-in only, and viewing as someone says so. On Access and connected apps, a key refused while viewing as someone now also says who you're seen as and who can make it.
Storage times on the Filing panel are right wherever you are. When the show's storage was set up and when it was last checked showed in the wrong time for anyone not on UTC. They now show your own local time, as the rest of Throughline does.
A seat marked sensitive again while it's being confirmed stays sensitive. On the Messages panel's held seats, if one coordinator marked someone's seat sensitive again while another pressed Confirm & provision, the new mark was lost and the person was invited anyway. Now, if the seat was marked again before Slack is asked, it stays held, nobody is invited, and the person confirming is told and shown the new reason. If Slack is already inviting them, they're seated and the seat stays marked sensitive. If someone chose Invite them as usual in the meantime, the confirm goes ahead.
Two people deciding one version at once reach one decision. If a department head approved a version at the same moment the director sent notes on it, both went through and the version kept whichever landed last. Two directors approving together released it to the whole production twice. Now the first decision stands, and the second person is told the version has already moved and what it is now. Two notes sent at once are both kept.
A department's weekly spend shows the right time wherever you are. On the budget pages, the time a week was submitted showed wrong for anyone not on UTC. In Los Angeles a week submitted an hour ago said "just now", and hovering over it showed a time seven hours out. It now shows your own local time, as the rest of Throughline does.
A script change approved twice at once lands once. If the same script change was approved twice at once, the revision was applied twice and the departments were told about it twice. An approval and a dismissal at the same moment could leave the change saying one thing and the pages another. Now the first decision takes the change and the second is told it was already decided, so the revision lands once.
A week the budget office has accepted stays closed. If a department filed its week at the same moment the office accepted it, the filing could land on top: the new numbers could replace the accepted ones, and a week could even go back to waiting for sign-off. Now a filing that arrives after the office has accepted is turned away with the line it always used — ask the production office to reopen the week — and the accepted figures stand.
Slack's times on the Messages panel are right wherever you are. When Slack was connected and when it was last checked showed in the wrong time for anyone not on UTC: seven hours out in Los Angeles. They now show your own local time, as the rest of Throughline does.
Delivery times carry their time zone in the API and over MCP. A delivery's sent and created times, read through /api/v1/deliveries, list_deliveries or the deliveries resource, now end in +00:00. Before, a client read them as its own local time. The same goes for when a person last signed in, in list_users.
The Slack panel only says Slack stopped taking messages when it did. If someone on the show had no Slack account linked, the Messages panel used to say Slack wasn't taking messages from the show, even though nothing was wrong with Slack. Their messages go to their phone, as the panel already says, and the warning now only shows when Slack really turns a message away.
Viewing as someone can't make a key or connect Slack in their name. Throughline support sometimes looks at a show through a crew member's eyes, and on a demo show anyone can. While viewing as someone you could make a Throughline key, match a Slack account to a crew member, connect your own Slack account as them, or connect a Slack workspace to the show, and each of those lasted after the view ended, in the other person's name. Now they're made from your own sign-in only; viewing as someone says so and makes none of them. Reading and revoking keys work as before.
A tool you ask the Forge for is yours, not your name's. Throughline used to know who asked for a tool by name alone. So someone else on the show with the same name could change it or roll it back, and read its note of what it doesn't do. Now a tool belongs to the person who asked for it, and keeps belonging to them if their name changes. You need to be signed in to ask for one. Tools from before are matched to their author where exactly one person on the show has that name; where two people share the name, both can still change it, as before.
Tell Throughline where scenes shoot. On the Script page, under Where scenes shoot, you can now say which scenes shoot somewhere other than the show's home base: the lake house scenes in Squamish, say. Those scenes read that place everywhere, straight away. A place set on one scene still wins, and a day with its own location keeps it. When all of a day's scenes shoot at one place you named, the weather is that place's, and says so. If two of your instructions name different places for the same scene, the scene says so instead of picking one. Say the same thing twice and it's filed once, and you're told so. Instructions you stop stay listed under Stopped, so you can read what you stopped.
A message that went out is never reported as failed. At a busy moment, a message to your department could stop after the first person, and a Slack message could tell you it hadn't reached people who already had it. Now everyone gets the message, you're told everyone it reached, and nobody gets it twice. If we can't finish writing down that it went, that's a note for us, not a warning to you.
The home base shows which time zone it will use. Throughline works out a show's time zone from the town's name, and a town can share its name with somewhere far away: "Santiago, Spain" was read as Chile, and "Lima, Ohio" as Peru. The new-show form and Show setup now say it under the place, for example "Time zone from this place: America/Santiago. Pick another if that's wrong." If a time zone is already picked, that one stands. On the new-show form, typing a different town over one you picked no longer brings the old town's country with it. On both forms, a home base name is one line of 120 characters at most. A show whose saved name is already longer keeps it, and can still change its spot on the map, its country and its time zone.
A Slack brief that doesn't go out is told, the same as a WhatsApp one. When Slack refuses one of Throughline's briefs, usually because the show's Slack connection has stopped working, the show's owner now sees it on Today, beside the WhatsApp line: "The Slack brief could not be sent. Ask Throughline support to help." It stays one dated line, not a new one each morning, and its date tells you when it last happened. Our support team's own view clears once a brief gets through. Someone we can't find on Slack isn't counted as refused: nothing was asked of Slack, so nobody is told. Our support team now sees what the provider said, under your show's name, so they can put it right faster.

2026-09-21

A new show starts with its home base, and nobody reads somebody else's city. Creating a show now asks where it's based — a best guess is fine, and the production can change it in Show setup later. The show's clock, its phone list and when Throughline stays quiet all follow that place from the first minute; if we can't tell the time zone from the place, the form asks for it. Shows that never said used to read "Vancouver, BC" on every unplaced scene. Now they say "No home base set", and the people who can fix it see a line on Today and in Show setup asking where the show is based. A scene someone placed by hand keeps its own place, and its page says which one it's showing.
The person who built a Forge tool can read what it doesn't do again. A tool can note where it falls short of what was asked for. Since the last Forge change nobody could see that note. Now the tool's author and anyone who can change Show setup read it, under "What it doesn't do" on the tool and on its build page. If a note needs tidying before it can be shown, you see a short line saying the tool works differently from what was asked, with a reference to quote, instead of a sentence with holes in it. Whether everyone else on the show reads the note is still being decided.

2026-09-20

A tool that didn't build says so in plain words. The Forge workshop panel used to print the raw output of the tool's own tests — a wall of machine text, sometimes a file path from inside Throughline or the name of the company whose model wrote the tool — and anyone on the show could read it. Now it says how the last build or change went, in the same words as the build's own page: the tool is live, its checks didn't pass, or the build stopped before it finished. It lists what's worth fixing, like how many boxes have no help text, and a change that didn't land tells you the version you had is still running. The reference is on the panel too, and Copy takes it with the rest, so whatever you send us is something we can look up.
A message Slack won't take is no longer recorded as sent. If your show's Slack connection had stopped working — a token pulled or expired — Throughline still said the message went: the sender saw it delivered, the record said the same, and a copy sat in the person's thread although nothing had reached them. Now a message Slack refuses counts as not sent. The sender is told who wasn't reached, the show's Slack card says the connection needs attention, and a message from Throughline itself is tried again on the next pass. A message that DID go out and only failed to be written down still counts as sent, and is never sent twice.

2026-09-19

The schedule stops citing a rule that doesn't exist. It used to say a "SAG six-day rule" decided whether an idle run is held or becomes a drop. There is no such rule. What Throughline knows is your own convention — how many idle days this show holds and pays. Your office can set that number on the schedule's settings; until somebody does, it is the one we ship with. Past your ceiling it reads as a drop, and a drop has its own conditions in the performer's agreement, a long enough gap and a firm recall date, which Throughline does not check. No figure on your board or your money card moved.
When a build or a change fails, Forge says so in words you can act on. It used to print the error exactly as the code hit it — an exception, a test runner's output, or the words the model sent back, on your screen and in the running log beside it. Now it says the build stopped, or that Forge wrote it and its own checks didn't pass, and gives you one short reference to quote if you tell us. A change that fails says your tool is untouched. The detail is kept on our side under that reference.
A standing rule filed under Transportation stays there. A rule like a unit move to the tarmac used to turn into a Picture Cars rule the next time we updated Throughline, and its item and its place in each scene moved with it, without a word. It now stays where you filed it. Only rules from before Picture Cars arrived on 10 September were meant to move: that happens once, and a show whose rules moved sees it in its activity. If you have an older rule for the same item as one you filed since, that one stays under Transportation too — one item belongs to one department, and you've already said which. If one of yours moved before this fix, file it under Transportation again.
The weather on Today is your show's, or there isn't any. Show setup saved the name of the place you work from and nothing else, so every show anyone had set up read Vancouver's weather under its own city's name. The place name is now a search: pick where the show is based and its weather follows. A show that has a name but hasn't picked the place shows no weather and says so, with a link to Show setup — we'd rather tell you than show you somebody else's afternoon. A day with its own location still gets that location's forecast. Change the town to a different one and the old town's spot on the map goes with it — type Vancouver over Atlanta and you get no weather until you pick Vancouver from the search, never Atlanta's afternoon under Vancouver's name.
If your show was left on two Slack workspaces, it goes back to one. A show that had the Slack simulator turned on over a connected workspace kept both, and anyone whose Slack account lives in the real workspace quietly got a text instead. The next time Throughline starts, the simulator is turned off, your activity says so, and those people are back on Slack. Shows on one workspace are untouched.
A message to your department reaches everyone, once it has started. If a big save was running while you sent a message to a department, it could stop after the first person, and you were told the whole message failed. Now it goes to everyone, and you're told who got it. A Slack message that went out is no longer marked as failed either.
A message from Throughline is no longer lost when it lands during a big save. A reminder or call-time change due to you while a long script breakdown was saving could vanish for good from your Throughline thread, or reach your WhatsApp with no copy in the thread. The thread message now goes out on the next pass, and nothing is ever sent to you twice. A WhatsApp text that goes out while Throughline is saving something big is added to your thread a moment later. If we still can't add it, you keep the text and we leave a note for ourselves instead of sending it to you again.
The assistant says what it does in one way. Its panel, the reasons it gives when it can't start, and the line it leaves in your activity all say the assistant reads the show. The show's Tools page says a script is broken down, the same as the Script page, and a script last broken down without intelligence now says so in the Script page's own words.
Reconnecting Slack to a different workspace asks first. If Slack hands your show a workspace other than the one it was on, nothing moves until you choose: move to the new one, or stay where you were. Nobody's Slack account changes while you decide, and a question nobody answers is undone after fifteen minutes. If you say no, the panel says so, and landing in the workspace you turned down asks again.
Times show in your own zone from the moment a page opens. A routine's next scheduled time, a delivery's due day and sent time, and the time on a feedback line at home all print in the zone on your crew card, then your show's main location. The small time stamps across the app used to show your browser's zone until you moved around the app. A vendor request's filing time and an engagement's start and end now carry their time zone too, so they no longer read an hour or more off.
When two people answer the same line of your note, the second is told. Their answers used to be taken without a word and saved nowhere. Now they hear that those answers weren't saved because the line is no longer on the card. Reloading the page shows where the note stands. A press that saves some answers says how many it couldn't, and pressing twice by accident is not an error.

2026-09-18

An answer to your old note's questions is no longer lost without a word. When your note is saved again, the old note's questions now leave the card at once, instead of when the new note has been read. A page opened before the save that answers one of them is told its answers were not saved and why, instead of seeming to save and then losing them. The questions a script raised stay on the card.
Saving your note no longer holds up everybody else's saves. We read your note as soon as you save it, and until that reading finished, every other save in Throughline waited and could fail. That included the questions a script is asked before its read, which were paid for and then lost. Your note is now saved first, and reading it holds nobody up. If the reading is interrupted, the card says your note hasn't been read yet, and we read it when you save it again or before the script is next broken down.
Answering questions while a script is being checked no longer loses that script's questions. If a script's first look finished while you were answering the card, its questions could disappear and the script would be read again. Now both are kept.
Answering a question about one script no longer undoes what you told us about your note. A "No, not this" on another script's question could erase your answer that the instructions cover your note, change how an older note is read, or mark a note answered that nobody had looked at — and each of those made the next read start over on parts it had already done. Accepting one of your note's lines later no longer erases that answer either. It changes when you answer that question again, or when you leave one of your note's lines out — and then your whole note is used, as the card says.
Breaking down the script is called that wherever you meet it. The progress card, the questions about your note, the done card, and the notices when it stalls, stops or can't start no longer call it “the read”. An instruction now shows the day it was made where you are, not the day after. And when a line from your note reads like a rule, you're only offered the choices that can be carried out: a line from your note changes when you edit the note.
Changing an instruction says what it changed. Adding one says how many scenes already on the board it went onto; stopping one says what came off the board; taking one off a scene says what came off there. Each also says what the next read does with it.
A question raised about another script no longer refuses your re-read. Before re-reading pages that are already on the board we check that the pages and the board have not moved under you. That check was also watching the questions a reading raises and the stamp left when somebody settles a card — so answering a question about a second script, even with "No, not this", could refuse a re-read of your own. What somebody actually agreed to still refuses it, which is what the check is for.
The reading that sorts your feedback now says which model does it, and you can change it. Feedback we receive is read before anybody looks at it. That reading had no row on the models page, so nothing said which model did the work. It has one now, above the show picker because it is ours rather than any production's, and the row says what will really run — including when that is the stand-in because we hold no key for the model that was chosen.
Two people starting reads on one show at the same time no longer lose one of them. Each read asks the script a few questions before it begins, and when two of those ran at once only one set of questions was kept — the other script's answers were paid for, thrown away, and asked for again on the next read. Both are kept now, and neither script is read for questions twice.
A question on a card is only ever numbered once. Each question we raise from a script or your standing note carries a short id, and that id is what your answer is addressed to. Numbering restarted at 1 for every reading, so a second script's questions could open on the first script's numbers — and a Yes from somebody with the older card still open landed on a question they had never seen. A show now carries its numbering forward, and nothing is renumbered, so a card already open keeps working.
The Script page puts breaking down the script first and keeps one list of what you've told us. Breaking down the script is the first thing on the page and stays there, running or not, with earlier attempts under it in Read history. What you've told us about this show is one list: a department rule and a line confirmed from your note both say what they do and what they cover, with your note above them as a reference and everything else one press away.
The same instruction is not stored twice, and two that read alike are a question. Confirming a card you have not changed no longer repeats what it says. When two instructions read alike we show both, and the words that differ, and ask — keeping one says which one stays, keeping both is one press and we do not ask again.
Starting a read stops opening everything you have ever approved. What you have already confirmed is folded away and counted; what is new, changed or still unanswered is open.
A read settles a same-person question it is sure about instead of asking you. When two names might be one person, the read asks again on its own and only brings you the ones it cannot settle. The same person at another age, or with a double standing in, is still yours to decide: that is two bookings, not one name written twice.
Pressing Re-read says what is happening and asks its questions right there. The pages are checked before the breakdown starts; the screen says so while it works, and anything that needs an answer appears on that same screen instead of ending the read and leaving you to find it. Nothing is recorded as a failed read for a question.
A question raised about your pages can be answered. Where it used to state a decision and offer only yes or no, it now asks it with the real alternatives, an answer that leaves it undecided, and a box for your own words. Whatever you pick becomes the instruction the pages are read against.
Answering a card answers the lines you saw. A question that arrived after your page loaded stays waiting for whoever it was meant for, instead of being settled by somebody who never read it.
A read that finishes clears the notices from the reads that stopped on the same pages. The Script page shows where this script's latest read got to and nothing older, so a good read no longer sits under warnings about attempts that stopped days ago. A stop on a different script stays where it is: one show's scripts do not speak for each other.
"Technical issues affected part of this read" now means something happened to that read. It used to appear on reads that ran perfectly, because of what the provider behind them does not report. It names the part of the read it is about, or it does not appear.

2026-09-17

We ask for the department in your show's words. Reading your note, we now file each instruction under one of your show's own departments instead of a general name for it. When none of them fits we leave it for you to choose rather than guessing, and an instruction we can't file is shown to you instead of dropped.
When a read stops, the page says why and when. Each notice carries the read's own words and its time, so two reads that stopped for different reasons no longer read as one. The notice names the part of the read that had trouble, and the lines beside it no longer report a clean result from a read that gave none.
Percep look briefs have their own allowance. HQ can set each show's limit separately from picture descriptions. The office can see how many looks were requested. Stored briefs stay available, and an unreadable answer is kept so retrying it does not pay twice.
A read tells you roughly how long it will take before it has started. Until the read has enough of its own progress to go on, the page says what reads of that size have usually taken, and switches to this read's own pace as soon as there is one.
A show with more than one script is only asked about each of them once. The questions we raise from reading a script are saved against those pages, and every script you have been asked about is remembered now — so moving between scripts and back does not bring back questions you have already answered, or read the same pages again.
The note card says which line it could not save. A line we filed under a department this show doesn't have now says so on the line, and offers the choice: pick one of this show's departments, or leave it as a note. Every other answer on the card saves, and the read can start. A line marked Not right keeps your whole note on the read, whatever you answer about coverage.
A failed render says more about why it stopped. A render refused for showing a public figure used to arrive as an unexplained stop; it now says a provider declined the content, which is something you can act on. A render interrupted by a restart is still not treated as a refusal.
A read now says when two entries would have one thing ordered twice. After the element list is finished, the read compares entries that share a scene and a word, and asks about each pair on its own: one thing, one thing at two points in the story, or two different things. Double orders come to the tray as the card you already answer, with both entries and the lines they came from. A pair you keep separate is not asked about again.

2026-09-16

HQ opens on Shows. Open a show's console directly from the list to see daily caps, script reads and document style. Analysts can read; ops and admins can make changes.
Saving how your show reads its script always tells you what happened. If there are rules to approve you go to them; otherwise the save says the note did not change, or is empty, or is being used word for word — and an emptied note says what still stands.
A read says once what held it, and your note is answered before the read starts. The questions tray states a condition that held the whole read at the top, with what to do about it, instead of repeating it on every card. Saving your note asks about anything it could not turn into an instruction, and the read waits until you have answered or chosen to have it read exactly as written.
The script is read once before the work starts, and you answer what it raises. Uploading or pasting pages, or reading them again, now asks the few whole-script questions only you can settle — a role that appears at two ages, a standing item the pages never name, a story told in two eras — on the same card as the questions about your note. Nothing is read in detail until you have answered them, and a script that raises nothing does not stop.
Script reads keep going after an interruption when new work was saved. The read says whether Throughline was updated or restarted. If two restarts happen without any new work, it stops and explains how to try again.
HQ refusals say who can help. A refused request explains what happened, which HQ seat you are using and who has the access it needs. An expired sign-in asks you to sign in again.
A read tells you how much longer it has. The script page gives one estimate instead of a wide range, worked out from how fast this read is actually going, and the line above the bar says how much of the whole read is done.
Unavailable reading choices are caught before the work starts. The Script page explains which setting needs updating and who can help. If an optional check becomes unavailable during a read, usable answers are kept and the page says what happened.
Script questions are easier to read and answer. Each card shows the question, why it needs your attention, the script passage and your choices in that order. Saved answers stay beneath their question, and Something else opens only when you choose it.
Add standing elements from the Script page. Script approvers can name an element, choose its department and say which scenes need it. Saving adds it to matching scenes already on the board. A refused save keeps the draft and explains who can help.
HQ can fetch missing script-read prices. The usage screen names figures priced from the provider's report and those that match its bill. Finished reads are checked automatically; the paste box stays available when a report cannot be fetched.

2026-09-15

OAuth discovery points to the current MCP documentation. The link now opens docs.criticalperiod.co/mcp. Existing API and MCP addresses are unchanged.
A re-read that is too long leaves its explanation. The office activity and Script page say why the read could not start, just as on a first upload.
Choose whether the reading rules cover your note. Yes means the next read uses your approved rules and any remaining lines. No or no answer keeps your whole note. Editing the note clears your answer.
A WhatsApp brief that could not be sent is visible on Today. A saved notice tells the show what happened and who can help. Support sees when sending stopped in HQ Readiness, until a later send succeeds.
See which character cues were not read. The scene page now lists these lines beneath the cast and explains why each was left out.
A refused Show setup change says how to move forward. The notice says what was refused, who you are on this show and who can help. The button comes back, and a successful save clears the notice.
Script read notices now appear on the Script page. About the script read shows what stopped or recovered, newest first, and updates while the read is in progress.

2026-09-14

These docs, and the addresses in them, now say Critical Period. The company behind Throughline is Critical Period, and the API and MCP doors are api.criticalperiod.co and mcp.criticalperiod.co. Every example here has been updated. The old leeme.co addresses still answer, so anything already pointed at them keeps working — but build new work against the ones above.
Invented quotations get their pages checked again. When too many quoted lines are absent from the script, the affected pages are checked again before anything reaches the board. If they still cannot be verified, the Script page explains why the read stopped and support receives the reason.

2026-09-11

A busy model no longer ends a read. When a provider says it is busy — which it does, in bursts, and usually clears in seconds — the read waits about a minute and keeps going, and the line under the bar says so instead of going quiet. If it is still busy after that, the read moves to a backup model where one is set for that stage. Only when there is no backup, or the backup is busy too, does the read stop, and it now says which model was busy and that the pages already read are kept — where it used to show a page of error text and tell you to ask the production office.
A script read now runs until it is finished. It used to stop itself ninety minutes in, whatever it was doing, and the line under the read counted down to that time. Long scripts were losing work to the clock rather than to a fault. A read that has genuinely stopped coming back still fails, in plain words, with the pages already read kept — but a read that is working is left to work. The line under it now says only how long it has been going.

2026-09-10

Names that look like duplicates are now questions on the review tray, and something reads them before you do. That list used to be worked out again every time the Script page opened, and nothing checked it — which is how a card asking to combine eight unnamed scene placeholders reached a producer. The names are compared when the script is read, the same second reading that looks at every other question looks at these, and they are answered where the rest are: combine into the name you pick, keep both, or neither, with your note. Keeping two people apart still tells the next read to keep them apart, and combining two spellings now sticks. A duplicate you make by hand between reads waits for the next read.
Before a read starts, the Script page repeats back how it read your note. You can edit a line, drop it, or keep the note as you wrote it. Hedged wording stays yours. A first read waits until you answer. A re-read with the same note skips that screen. Until you confirm a line, we work from your note exactly as you wrote it.
Picture Cars and Production Design get their script work. Vehicles on camera go to Picture Cars. Sets, backings, signage and graphics go to Production Design. Existing vehicle entries move to their picture-vehicle owner; crew transport stays with Transportation.
Saving a brief whose last reading failed reads it again, even when the words did not change. The page says we read it again. A save of an unchanged brief that last succeeded still writes nothing.
A second reading that quotes the script now does the bookkeeping. A split, a fold of two of the same thing, taking something that is not an item off the list, citing the line the reading named, moving department, or putting an item on the scenes its own line names — those leave a receipt you can undo. Two names for one person still ask. A line the page never said still asks.
Sunglasses, gloves and an N95 sit on the costume and props lists both. Costume establishes them; props handles them on set. The Script page no longer asks which department owns that kind of item. A passenger mask or anything else that is one department still asks.
A count the script never gave is left blank. Carry-ons and the like stay unquantified. Nobody is asked how many. A number the page does give is kept.
Saving how this show reads its script now says if that note could not be turned into questions. The brief itself is kept. The page says so, and to save it again to retry. The standing questions stay until then.
A script longer than we can read whole is refused, with the page limit. We never read an outline instead. Nothing is read, and the Script page says so.
The re-read dialog says whether these pages are already read. When a matching read is on file it says so, and that only changed pages would be new work. When nothing matches, it says this read is new work. The count of answers that carry over is still there.
The element check says what happened to its objections. The Script page separates corrections, questions for the office and changes that could not be applied, with reasons. Older reads say when those outcomes were not recorded.
A Pulse send that fails does not stop the rest of the show's tick. The failed send is recorded. The next person, and the next kind of notice, still go out. The Pulse board and the show's HQ Pulse tab name the channel, the last error and the time. The line clears when a send on that channel gets through. A simulator is named as a simulator — the same words the rest of the product already uses.
A re-read that stops after the pages are on the board still offers the tray. The Script page says the pages are on the board and the questions on this read may not have been assessed. A failure before the board is saved still says to ask the production office. Show setup → Tools names the cause for staff.
The Script page keeps answering while a script is being read. A long read used to take the whole database for itself for seconds at a time, and anything that asked for the state of the read in that window got an error instead of an answer — four times during one script. Reads and writes no longer wait on each other, so the page keeps telling you where the read has got to while it is still going.

2026-09-09

Show setup now has a main production location. The phone list and address search use it. Empty until the office sets it. A first scheduled day can hint at a place; it is never filled in on its own. The onboarding form says which show location it is using. Address search looks there first, prefers a house and street when you start with a number, and names the region it searched. A country without a stored map box still narrows the near search to that country.
A show's paperwork now wears that show's look. Throughline staff set it from the show's HQ page — the printed name, a mark, the colours, the type, and the line at the foot. A show that has set nothing already uses its own colour and the house type. The next breakdown, crew Day Out of Days, and turnover lineup pick it up; paper already sent stays as it was.
When too many questions on a tray would not get past a professional, the Script page says so. The line is the parse team's: N of the M questions on this tray would not get past a professional; the parse team has the list. It sits under what the read landed, only when that is true.
Script review can answer what the pages already settle. With the review lane configured, supported factual answers become receipts under Answered for you, with Undo. Questions about meaning stay for a person. Different names are never silently joined, and nothing is removed from the board. The review's cost and counts are recorded; its Script-page summary and live calibration are still to come.
Library file tiles load a small picture, not the original. When a file has a stored thumbnail, the asset grid, the ingredient browser and the canvas fetch that. If the filespace picture is still on its way, the tile says Preview on its way instead of stretching the one-pixel stand-in.
Every script read now uses intelligence. Uploads, revisions and re-reads all get the full read: the cast, the questions and the second reading. A show can still switch intelligence off for the day-to-day features it uses once the script is on the board. The script itself is too important to read any other way.
A show with no model connected is told it cannot read a script. One line, in the same words on the Script page, in the activity feed and on the read itself: add a provider key under Keys in HQ, or ask us to set one up for the show. Nothing is read and nothing is charged.
A script read before this change says so, and can be read again. The Script page names the date it was read without intelligence. Reading it again is the office's click, priced like any other read; it never happens on its own.
HQ's Script intelligence tool says reads always use intelligence. The switch is for the day-to-day once the script is on the board — today, the methodology brief's own reading of the note. A show with no provider key is told it cannot read a script, in the same words as the Script page.
A running read names when it started and when it will stop itself from the first look. The remaining-time line says it is an estimate, or that the first stage is still setting the pace. It does not say the software is working the clock out. The stages speak in production English: Discovering characters, Matching names to people.
A script read stays on the Script page when you come back. Another browser, another device, or a sign-in that cleared this one still shows the running read, who started it, and when. A read that finished while nobody was looking lands the next time anyone opens the page.
Show me the twelve opens one card per place. The folded scene question becomes a receipt — Expanded into 12 cards — under Answered for you, with Undo. Each place is an ordinary scene card with its page and scene, and Break it out or Leave it inside.
The Script tray groups element questions by what they ask. Which scenes, something the list is missing, one item or two — each kind is its own sheet, so you answer one kind in a pass. A department question is a picker. A scenes question is tick-boxes. Evidence is the script page, with the scene and page above it — never a line number. The list of lines judged not to be characters names the scene and page the same way. A scene the second read already took off sits under Answered for you, with Undo to put it back. When a spelling of a person was already settled, that receipt sits on the open card as its own line, so you can see what was answered without opening a closed row.
Review evidence reads as a script page. Housekeeping cites scene and page — Sc. 22, p. 47 — never a line number. The quoted lines sit in Courier like the page they came from, with a line or two of context, and the cited line marked. When the saved draft has no page breaks, the card says so.
The second reading can make a supported correction and show what changed. A department move, a named missing item, a scene taken off an entry or a duplicate folded into another comes with a receipt and an undo. Undo puts back the saved entry, including its evidence and scene links. If someone has edited it since, the card asks for review instead of overwriting their work. Splits, removals and quantities stay questions, with the reading's proposal named.
Element cards answer the question they ask, and the obvious ones are already answered. When the second read says a scene rests on a line that does not describe the item, the scene comes off that item and the card becomes a receipt: it says what was taken off, and one click puts it back with the line it was cited to. The item never leaves the board — if that scene is the only one it has, the card asks instead. Everything else asks with answers that fit the question: one item or two names both items, which department lists the show's departments with the current one first, how many, which scenes to tick.
The review tray asks once per person, not once per spelling. Every card about the same person is now one card, whatever each was asking. A name the script spells two ways is merged and shown as a receipt you can undo. Two different names are never merged for you: that one always asks, and nothing goes on the board until you answer. And the places a script cuts to without a slugline come as one question listing all of them, each place once, instead of one card apiece.
A nickname three people hold is three questions, not one. Each holder is asked whether that name is theirs. The other people stay on the card so you can see their cues; answering one card does not decide the others.
One person reaches the tray as one card. When the read finds a gunman and a named reveal that are the same role, it asks once, with the line that says so: Same person or Keep separate. It never folds them on its own. A second card about a name that already has one folds into it. When that extra look ran, the Script page names what those calls cost, next to the rest of the read.
Connected tools can put a recent combine back. A script approver can undo a saved combine for 72 hours, provided its attached work has not changed. The activity feed records the result. The housekeeping button is still to come.
A recent combine can be put back for 72 hours. Housekeeping's confirm names that window, and Put it back reads the same offer the activity line carries. A scene, booking, or later combine that already changed the work is refused in those words.
A line of action is not a person on the cast list. If a read uses a sentence as someone's name, they stay on the board as Unnamed until you say who they are, and a card quotes the line. Real names — including ones like ATC (LA CENTER), and a slash of two names — stay as they are.
Combining character names brings their props and wardrobe with them. Items named for a character take the surviving name in scenes and department lists. Searching the old wording still finds them.
Decisions the parse already made sit under Answered for you. They do not count in the number at the top of housekeeping. Each one says when it was applied and has Undo. Accept all keeps them in one click, using whatever answer that card already offered. If one cannot be kept, the rest still are, and the tray names that card by its headline and says it stays there.
An interrupted script read gets one automatic retry. After a server restart, the Script card says the read is carrying on. Answers already saved are kept. If another restart interrupts it, the read stops and says so.
On housekeeping, an AD can combine two people even when the list didn't propose them. Pick the names, choose which one stays, and confirm. Every scene, image and file that pointed at the others will point here. The other names stay as other ways of writing the same person.
A stall dump is readable from HQ. When the event loop stalls, or a script read captures its stacks, the file lands in Diagnostics — staff sign-in, newest first, click to read. A name that walks out of that folder is refused in words. Files older than fourteen days are dropped at startup.
A one-line schedule from the AD's own software now comes in. Yellow and Goldenrod one-liners are read as one-liners, not turned away as scans. When the heading is glued to the first name on the list, that person is still on the list, and the scene still reads LEO / STELLA. A file that really has almost no text is still called a scan. A file that has text but isn't a call sheet, one-liner, shooting schedule, or DOOD is told that, without the guess.
If a show code is already taken, the refusal names the show that holds it. Creating Copper Hollow as DD when an earlier Copper Hollow already has that code now says which show, and whether it is on the air or wrapped. A wrapped show can release or change its code from Show setup; a show that is still on the air cannot, because mounts and folders use it. A code that was assigned from the title on an older show says so on that page, so it is not mistaken for one somebody chose.
Restarting a script read no longer doubles its bill in HQ. Saved answers count once across attempts, including failed reads. Money spent on calls that were not saved still counts.

2026-09-08

The Brain panel says what the last automatic read actually did. A skip or a stop at the day's limit shows the same sentence the activity line does. A finished read still says it read recently.
Two spellings of one prop no longer knock the staying one off the sheet. If one record is leaving a scene and the other stays — same department, labels that differ only in capitalisation — the one that stays keeps its row. When an item moves onto a scene that already has it, the notes, the quantity and the lines it quoted travel with it rather than vanishing.
Opening material added or changed in a later draft waits for AD approval. Its script change names the source lines and asks about numbering. Approval keeps printed scene numbers and a changed opener's existing scene identity.
Two wearers no longer hide a question about two different objects. The Script result names dismissed per-person split questions. Duplicate entries and names that need an owner or description come to the review tray as their own questions.
Two people can answer different sections of one review card at the same time. Both answers and both notes stay on the card. Asking later or handing the card to another department also keeps the answer someone just gave.
A set-dressing item stays on a scene whose heading names that set. If the kept quotes are from another scene of the same set, it is on both. A costume that belongs to a character stays on a scene that character is in. Quotes from a different set, or from a scene the character isn't in, still move it.
A script read you start still runs when a spending limit cannot be set. The Script page and the activity line say so. A read nobody is watching does not start, and the message says how to run it yourself or set a limit in Show setup.
The assistant's timer no longer re-reads when it cannot read today's spending ledger. The activity line says so. A re-read you start still runs.
A name spoken over the P.A. is still that person. A delivery note in parentheses after a name — on the P.A., on the radio — still puts them on the cast list, including the cold opener.
A script read no longer stops on the first call when the text model was retired. The message says which model to pick.
Nothing is filed under a scene your board does not have. A read could hand back a bracket, a comma or a paragraph of its own notes where a scene number belongs, and items rode in on it — one read had 114 scene keys for a 97-scene board. Those replies are dropped now and the read says so, with the item count, and the commit refuses them a second time. The cold open is a scene, and it is kept.
A script read that had to be started again still shows what the whole read cost. If an earlier try stopped mid-way, the Script page names the money already spent, not only the last try — including a read that finished before this did. The spending limit counts that same figure, so a resume cannot spend past what the office approved as if the earlier try were free. The last try is named beside the total. Each part of the read shows what it actually spent next to what it was quoted.
A camera direction is not its own scene. A line such as EXTREMELY CLOSE stays inside the scene it belongs to. A jump that names a place — one already on a slugline in these pages, or a word like cabin, hall, or deck — is still its own lettered scene, and it keeps the parent's INT or EXT and time. A time of day left on a slugline (a moment later, minutes later) is not a place, and neither is shot. If a camera line was already split out, the next re-read puts it back, and its items go with it.
A prop that turns up in five scenes is one item with five scenes on it. The read used to pick one scene for it and the whole list sat there. Now it reads the lines it quoted and adds the scenes they come from. It never takes a scene away on the strength of a quote — a read only quotes an item a few times, and a hero prop is in far more scenes than it is quoted in. Where not one quote lands anywhere the item was filed, the quotes win and the item moves. And the pile the read could not place — items attached to the last scene it was asked about, quoting a line that never names them — comes to you as a card asking which scenes they are in. The card is a question, not a removal: nothing leaves a department's sheet because the read was unsure.
Flynn Bower is one man, not three. When the script writes somebody's name three ways — FLYNN BOWER, SPECIAL AGENT FLYNN BOWER, FBI SPECIAL AGENT FLYNN BOWER — the read now sees that on its own and puts them together, instead of asking you three times about three people. A title on its own never gets folded onto a name, and if two different people really could answer to the same name you still get asked.
A cockpit annunciator is not on the cast list. Machines that talk — an annunciator, a collision alarm, a lift voice — come to you as a question about production sound instead of arriving as a person to book. It is still a question: an annunciator can be a real voice role, and if anybody in the script talks back to it, it is a character.
Whether somebody is a voice session or an on-camera day is decided by what most of the read says, not by whichever reading came back first. One odd reading used to be enough to move a role between those two.
A card no longer offers you the wrong person first. If the script never writes that name for the person the card is about, the person it is written for leads the answers. The buttons are exactly where they were.
The person who creates a show can run its first read. The role they typed is their job, not a lockout. A show whose creator was not its administrator is fixed at startup.
The line under a running read uses your crew-card time, not a UTC stamp the browser guessed at. On a show with intelligence on, before the stop time is known it says it is still working that out. A show without intelligence, and a run that truly has no stop time, still say No deadline. Next fire times, a delivery's sent stamp, and the home feedback clock use the same zone.
Pressing Break down these pages says the note is being read. Until the save comes back, the button is off and reads Reading your note. A line says the note is being read before the pages are. A second press in that window does nothing. If the save fails, the reason is above the buttons and you can try again.
Costume opening Mara from Costume's workroom sees Costume's starting points. Other departments that have published one for that item are there to pick. Applying theirs adds their flow and leaves who owns the item alone. A starting point for a character will not apply to an asset, and the workroom list only offers starters that match what you picked. On the workroom, Apply a starting point applies the one you chose, or the button itself says why it will not. The list without a workflow opens on the work that department owes. A scene on the board that has none of this department's elements is not this department's work. Everything of this kind is the wider list.
A revision that changes a prop or a line deep in a scene files a card. A byte-identical re-upload files nothing. The first upload after this lands will surface every change the old opening-only compare had been swallowing.
The assistant now reads what the show is working on, not the first pages of the script. On a big show its picture is too small to hold everything, so it starts with the scenes on this week's shoot days, then the scenes changed in the last fortnight, then the rest in script order. What the assistant has read says what would not fit — "the script: 31 of 182 scenes" — instead of only saying that something was trimmed. The day the unit is on, its general call and the scenes up, now travel with every conversation. A short show reads exactly as before.
Ask the assistant about a day by its number and you get that day. "What's the call time on day 12?" answers with Day 12's call time and Day 12's scenes, from any screen, with your department's items on it. Ask about a day that isn't on the schedule and it says so, and says what the schedule does run. Before, it answered about today.
Every show with Intelligence on gets a second check of its element list. Nobody has to find a switch for it: the read goes over the props, wardrobe, and set-dressing list a second time and questions anything it cannot stand behind, the way it already does for the cast list. Anything it questions comes to you as a card rather than quietly changing the sheet. A show that has turned the second check off keeps it off.
The first line of a revision card names what was added and taken out. The crew-readable sentence follows it. Marking a change as having no impact is a named decision on that card, not a rewrite of the line.
A link that names a version opens that version. Pulse and the workroom Review and With the director cards carry it through. If that version isn't available to you, the page says so in one line and shows the current one.

2026-09-07

A show that asks for a second check of its element list is quoted for it. The estimate before a read now includes that check, it is named among the parts a changed brief would re-buy, and the limit the read stops at is worked out from a figure that has it in. A show that has not asked for the second check sees exactly the estimate it saw before.
A stalled read is a card, not a raw error. If a script read has gone quiet, the Script page says so and how long it has been. Stop this read is there for an AD, a producer, or an admin. Anyone else sees why they cannot. Starting another upload while one is still running opens that same read.
A runner cannot start a paid script read. Upload and paste ask the same people a re-read already asks: ADs, producers, or admins. The job names who was signed in, not whoever the form typed.
An unheaded opener now raises one question, not two. If a second undeclared-scene suggestion points into the same opening lines, it is folded into the opener's card. A separate mid-scene question now names the scene it would split and the lettered scene it would create; if the line cannot be placed in a scene, the card says so plainly.
Show setup → People can keep someone out of automatic invitations. Crew who can change that page pick the person and why. They will not be added to channels on their own; someone still confirms the seat by hand. Clearing it invites them as usual again.
Show setup can change the rest hours, the working week, a department's details, and how episode codes print. People who can change those pages do that there. The week and a privacy change show what would happen before you apply them. A department's default calendar names the workflows that still follow it, so the first look is every follower, not none. If the show moved since you opened the page, it says so and asks you to reload. A department head can still turn Workflows on or off for their own department. The show code stays put.
Show setup can change the title, the phase, the colour, and the clock. People who can change The show do that there. The show code stays put: mounts and folders use it, and changing the title does not. If the show moved since you opened the page, it says so and asks you to reload.
Saving how the show reads its script now asks whether we heard it right. Each reading is Yes, No, or in your own words, with the recommended choice marked. Nothing typed is written until you confirm it. The note keeps riding the read until you say the answers cover it. A note about who is who, said in your own words, rides the character matching — not the element read — and shows next to those answers the next time you open the interview.

2026-09-06

Re-reading pages now shows last time's answers, and says what a changed brief would re-buy. The interview on that door is already filled in. Save with nothing changed writes nothing. A booking change re-buys the character matching, not the whole read; a show that never saved a new brief keeps the read it already paid for.
A new show starts ready to read its script. HQ shows the show's code as it creates it and starts script intelligence on for both a single story and a season. The production office can switch it off or on again from Show setup → Tools, which also says whether script intelligence can run right now.
Your settings and Show setup speak the house voice. The page is You. Slack, connected apps, memory, and what the assistant has read are said in production English. A no on a reach is still said once.
Reading access sits on Show setup → Access. On a series, who may read ahead is next to the keys, only for someone who can change that page. Your settings no longer points at Vendors.
Your settings is yours. The apps you've connected show once. Saying no to a reach is said once, at the foot of those cards: Throughline will do less on its own, and less for your department's team through you.
A scene whose elements were not read now shows on the tray. Ask again reads that scene's elements only. Leave it empty keeps the scene without them. A long scene on the one-scene-at-a-time path is asked in parts rather than coming back blank.
A parse now shows one line, a bar, and about how long is left. The button stays Break down these pages. The same job no longer reports itself three times. The bar does not sit at the end of a stage while that stage is still the current one. When the product already priced the read, the time left is a range in minutes, not a false countdown. You can leave the page; the read carries on. When it finishes, that same block names what landed, what it cost, and how long it took. Before the drop, a show with intelligence on says so, and that a feature takes about an hour. How the read works sits on that block, while the read runs and when it is done.
While pages are being read, the progress line keeps moving as each stretch finishes. It no longer sits still until a whole batch of answers is saved. You can leave the page; a long stretch no longer looks frozen.
How the read works is a page you can open from the parse. Seven short chapters: the pages arrive, who is in each scene, who is who, the check, what each scene needs, then it is yours. It opens when you're signed in.
The show's controls now live on Show setup. People who can change the workspace, the look, the filespace, the keys, the assistant's brief, or the Day Out of Days do that there. Your settings keeps what is yours: your look, your Slack connection, your connected apps, your memory.
Dropping pages on a show with intelligence on now asks how the show reads its script, before the parse starts. Three questions: namesakes booked as two people, one performer across roles or ages, and anything the pages will not say. Remaining room is visible as you type; past 12,000 characters on the standing rules, or 2,000 on a note, the save is refused rather than trimmed. Skip is one click and is remembered for that revision. The Script page keeps the standing brief on a card, with Edit. The brief still reaches every prompt the way it already did.
The figure before a read is priced on the lanes that will actually run. A first read says so and gives a wider band. When the read finishes, that figure is closed against the bill so the next one can learn from it.
A jump to another place inside a scene is now its own lettered scene. The parse files it before anything is written: 59 stays 59, the cockpit is 59A, and each part has its own eighths and cast. An action line that just ends in dashes stays in the scene, including a longer grab that is not a place name. A card can still put two units back together.
The Director now sits in Direction. Every show has that department, open, with the Director as its head. A cinematographer retitled Director is still not the Director. Handing a Direction step to a vendor is their own department's call, and still not another department's.
Show setup is a place of its own. People who hold a job on the show can open it and read the values they already may. Money, access, and vendors stay closed unless that job includes them. Settings in the rail is now your own page.
A Producer keeps the Director and Producers look after they can read the office's paperwork. Being able to see those records does not take that look away. A Line Producer still does not get it; the Director still does.
The Director's say is a written appointment. A credit that says Director, or a cinematographer retitled as one, does not grant it. The show's Director holds it everywhere on the show; an episode's Director holds it on that episode. Being an administrator is not being the Director.
A show that already existed when we started writing chairs down gets those chairs at startup. A Producer whose credit is exactly Producer can sign a Producer step again. A Head Production Designer is still not the Production Designer, and a chair that was taken away stays away. Fresh demo worlds keep the chairs they were seeded with.
Handing a step to a vendor is the department office's call, written down. A person given that job on the department can do it even if their credit says something else. An Associate Producer's title is not enough on its own. The people who get told about a department's steps are that same written roster, and only while they are active.
Who may change an approval room, say where a scene shoots, or re-file a scene onto an episode is written down on the person, not guessed from the credit. Holding a pass to read the office's paperwork is not enough to staff episodes. The assistant will not grant a permission; it names who to ask. Who signs a step named for a chair is written down on the show the same way: an Associate Producer is not the Producer, and an empty chair stays unsigned until someone is named.
The automatic re-read pot is $15 a day, and raising a limit stays. An administrator changing a show's daily figure is a permanent change to that show, not an increase that dies at midnight. The $50-a-day ceiling is unchanged. The settings record names the old amount and the new one.
The sweep that writes today's permission list down will not guess. Two people on one show who share a name stop the run before anything is written. A grant that cannot come back because it was revoked is named as blocked, not applied.
A 2nd AD and a production coordinator can now edit and publish the shooting schedule. The same written roster covers the paperwork tray, Pulse runs, the egress audit, and the weather watch. Who may do that is written down on the person, not guessed from the credit. A Producer's Assistant still cannot decide script truth unless someone has given them that job.

2026-09-05

The assistant's picture of a parsed scene names each person once. When the sheet has several cue variants for the same person, the read the assistant uses lists that person once, in the order they first appear. Seed shows without a resolved cast are unchanged.
An unheaded opener now asks the production office how to file it. Its card offers scene 1 with renumbering, an A-prefixed scene without it, or no scene; dialogue and shootable action lean scene 1, while a prologue, pure screen context, or any existing shoot day leans A. The setting only preselects an option — an approver must choose it — while cast, elements and mid-script lettered splits keep their existing behavior.
Tiles now load a small picture. Library cards, the canvas rail, the ingredient browser, shot plates, review strips and workroom covers fetch a stored thumbnail. The full file is served when you open it, download it, or send it to a model. Files that do not have a thumbnail yet still show the original.
A re-file plan no longer moves empty placeholder renders into a task's folder. Only the July stub signatures, and only when the file is under 100 bytes, stay on the plan as placeholder files, not moved — a small colour sidecar still files. A census script lists every version or library row that still points at one of those stubs — on disk or on the filespace — and a confirmation-gated remove deletes only what that list named. A filespace copy we cannot open is listed as unread, not removed. Filespace deletes go out as agent work; Throughline never touches the mount.
Re-reading the show now has a real daily allowance. Each show gets one pot for re-reads started by an administrator and a separate pot for automatic refreshes, reset at midnight where the show is based. Settings shows what has run and what remains. An administrator can change either amount; if a read will not fit, it stops before anything is sent and keeps the existing brief.
A vendor's delivery now arrives whole or not at all. Throughline checks and hashes the private upload, claims its retry key before writing any shared staging file, then stages and verifies every byte. A step-bound submission commits the delivery and its sign-off together before publishing the complete directory. After a restart, Throughline settles a complete or already-moved delivery, leaves an incomplete one for the same-key retry, and names genuinely missing bytes Arrival failed; none is shown as a ready file.
Vendors now have their own focused workspace. A vendor sees only steps both assigned to them and inside their live access, can leave progress notes, and can submit files for production sign-off. Their pages do not expose the production's navigation, approval buttons, date controls, pins, or internal notes. Until 1 December 2026, older submission clients that omit an Idempotency-Key still work with a dated warning; after that, the key is required so a retry cannot create a second delivery.
Who can change the show is a little tighter, in six places. A Producer's Assistant can still see colleagues' contact details, but cannot edit the crew list — that's the production office. Only an administrator can re-read the show for the assistant, or make a standing routing rule from chat (the same as changing it in Settings). A connected app can invite people and change a role, but it cannot name a head of department. A tool turned off on the Forge now shows Turn it back on, and rebuilding one is an administrator's.
The office is no longer told a read failed when it stopped at its limit. A script that hits its spending limit or its time limit still stops, and the board is still untouched, but the activity line now says it stopped — not that it FAILED. FAILED is what you retry; retrying this one spends the same money again.
A script read now stops at a spending limit instead of running away. Every read is priced before it starts and given a limit based on what reads on that show have actually cost, with room to spare. If it reaches that limit it stops and says so plainly: nothing on the board changes, and the pages it already read are kept, so running it again doesn't pay for them twice. A department second look works the same way.

2026-09-04

Provisioning a show's storage no longer waits on the click. Settings shows each step as LucidLink builds the filespace — creating it, waiting for it to come up, building the folders, setting up department access, minting the storage key — then Ready. A second click while that's happening is refused until it's done. Tearing storage down is still the short wait it was.
An item both readings of the script found is marked, and sorted to the front. When the breakdown reads the pages twice and both readings list the same item, the scene shows a quiet "Seen twice" next to it, and those items come first in their department and first in the review tray. It is a note of confidence on something already right, not a problem to go fix. Pages read once look exactly as they did before.
A film cannot have two scene 14s. When a later draft wants a number that's already taken, the tray asks: file it as 14A, or say it continues the one that's there. Neither leaves the lettered insert for now. A show that already has two scenes on one number still boots; those are the questions waiting.
A breakdown now records what it cost, including one that failed. Every lane the parse used is counted, not just the first, so a run on two models is not reported at a fraction of its price. A parse that stopped partway still shows the spend up to that point — the money is gone whether or not the answer arrived. Running the same pages again after a failure adds only what the retry actually buys: answers that come back from the bank are free, and they do not erase what the first attempt paid.
Re-read these pages without losing the board. A script approver can preview which scenes stay, which are new and which are no longer in the draft, then confirm. Shoot days, approvals and hand-set shooting locations keep pointing at the same scenes. Human answers and standing element rules carry over. If the pages or the board change after the preview, it asks you to look again.
When two departments name the same thing, the tray asks once. The card lists every department that proposed it. Yes puts a line under each of them; one department books and pays for it. A department can say it is not theirs and hand the card to another — including one that never proposed — and a department that already passed cannot take it back. If the last named department steps off, the question goes to the producer. Saying it is not an element still applies to the whole show.
A standing element rule now lands on every new scene, not only the parse. A split, an approved revision, a parse with the intelligence pass off, and a re-parse that reused the last answer all carry the element. A scene somebody took it off stays without it on those paths too.
A later draft that rewrites a scene's location line keeps that scene. When the pages and the board both carry the script's own numbers, the match is the number — not the slugline. A reused number over different content still comes in as a new scene. Older boards, and boards numbered by position, keep matching the way they do today until a full parse records how they were numbered.
A parse that cannot read a cue-shaped line as cast now says so. The scene and the activity line carry the count. The cue pattern is not widened; the drop is no longer silent.
A department's second look waits its turn, reads the show's own brief, and does not let an untouched card hide the next draft. One look runs at a time. An omitted scene no longer blocks a live proposal. A card nobody reviewed is replaced when the next look finds the same thing; a card someone put off stays put.
A later draft that only numbered some of its scenes no longer asks you to renumber the ones that caught a margin number. Those numbers stay put until the draft numbers every scene, and each one distinctly. A new scene that wants a number an omitted scene still holds is filed as a lettered insert (12A), so two scenes never share a number. A titled name on a later draft — Co-Pilot Dace beside a Dace you already have — lands on that person, with the other spelling kept as an alias.
The review tray names the read that produced the questions on screen. A later shadow run or a second look no longer steals those numbers. When the cards come from two reads of the script, it says so. The discarded-lines list belongs to that same read.
The schedule office can publish the whole show and keep the crew's copy as the comparison line. Preview names added, removed, renumbered and moved days plus scene reassignments. Marking a publication sent requires its real delivery receipt; the Days board then shows “was” beside dates and scenes that changed. An internal draft does not replace what crew were told, and a withdrawn sent copy is corrected by a new publication rather than erased.
A file card shows where that file has been. The latest move sits under the folder — from the old folder, or that it left — and the rest of the list opens from there. A file that has not moved looks the same as it did.

2026-09-03

An old Workflows filter link shows the full list, not an empty one. A saved In progress — or any word that is not Late, At risk, Ready, On track or No need-by — is treated as no filter. Late and Ready still filter as they do.
A look brief stays with the still it describes. Opening the same Released version after a restart does not ask for it again, and the box no longer holds every brief it has ever written in memory.
Moving a file on show storage keeps it the same library file. Its versions, tags and notes stay with it, and its record shows where it moved from and where it is now. A file that disappears waits 30 minutes before it is archived; if it returns later, the same record comes back.
A version tile loads a small stored picture, not the original file. Opening the version still shows the original. Worlds and seed media keep serving as they do.
A vendor submission that does not finish no longer hands its folder to the next one. New work lands under a name that cannot collide with a rolled-back row; existing folders stay where they are.
Asking the assistant to approve everything only talks about cards you can actually sign. A step sign-off still happens one at a time on the step. If nothing is waiting on you, it says so and stops there. A stand-in signature on the item names who stood in and for which role. The step's approval line opens that approval, not the whole list.
While the show's approvals are paused, a step that already has a sign-off card cannot be sent again. A decision that already landed still writes through.
A workflow row on the department list is Late, At risk, Ready, On track or No need-by — not a second status beside that. The cell is the same sentence the item header reads, so a past due never sits next to In progress. A missing need-by ranks ahead of On track, because an open date is less known than a flow that still fits.
An outside signature now belongs to one named person and one Approval step. The schedule office can bind that signer for a set time and revoke the binding later. Their approval key can Approve or Request changes on that step, and they can take back their own signature with a reason. A general write key and a name typed into the request can no longer sign for somebody; a turnover vendor lane cannot double as an approval lane. On Approvals, a workflow card now names the item and step it governs and links straight back to the work.
A workflow's Late / At risk / Ready / On track now comes from one reading of its steps, not a separate summary. Past a due date it names how many steps are behind; a raised hand or a last step that no longer fits the need-by reads At risk. A workflow with no need-by says so, rather than On track. An item reads the worst of its workflows, so it never says Ready while another department is still working.
A kept element that names a scan id now shows the element's name on the sheet, not the id. If the id has no name, the keep is dropped and the parse records which id and which scene.
Workflow notices stay off until the office turns them on. A sign-off waiting on you, a date that moved on work you own, a cascade that touched your department, and one daily digest — that's the set. Quiet hours hold a ping until morning rather than dropping it. Turning the notices off leaves a send that's already going out to finish, and doesn't dump a backlog the next time they come back on.
A published workflow can catch its items up, on purpose. The Workflows tab says how many sets still sit on an older version, and Update the N shows every change before anything writes. Fields a person set stay put; a step the template dropped stays unless you retire it. Put the update back from that same place. On the item, each field says whether it came from the template or was set here, and reset writes the template's value back.
One scene that cannot be read no longer ends the breakdown. That scene stays empty and the rest of the parse continues. If a second reader already finished, its answers are kept.
Saying where a folder belongs moves the files there. On Folder questions, That’s right puts the media with the named owner when that’s a different person or item. Leave it here files it under the parent it’s already in. Belongs somewhere else lets you pick any official home and moves a whole selection there at once. The files wait a minute so you can still cancel; they never go through the re-file plan.

2026-09-02

A workflow's formal sign-off and its Approval card now stay together. Sending the step for sign-off opens the show's Approval; Approve, Request changes, Reject, reopen, or unapprove there updates the step in the same save. The named role holder, delegate, or recorded stand-in can sign — being an admin is not a signature — and a signed step links back to Approvals with the deciding hand and date. Reopening keeps work that was already released moving while the department decides what to redo. Asking for that approval again starts from its Approval card, so the step cannot move behind it; a withdrawn submission leaves no false “waiting on you” card on Today and returns there when it is sent again.
A department can write its own workflow templates. On the Workflows tab — not a new tab — the department lists the starting points it already uses, edits the steps in a list with a live diagram beside it, and publishes only when every problem is named. A coordinator or producer in that department, the head, someone they've named on the contact card, or an admin can edit. Anyone on the show can save a real workflow as a template; if they can't publish it, it stays a private draft until the department does. Lead times copy as they are set. Taking them from this item's dates is a separate step that shows the numbers first. Nine steps is about as long as a workflow stays readable — more than that is a warning, not a refusal. If two people save at once, the second sees what it says now.
Putting a workflow change back is only offered when it can happen. Starting a step still puts back. Applying or replacing a starting point, or customising the steps, says so on the trail instead of showing a button that always refuses. Replacing a starting point now names the one that was there.
A department can take its own second look at the script. Choose which departments should look again after the first breakdown. Anything new comes to the review tray with a quote checked against the page; it doesn't change a scene until somebody says yes. A repeated item, or one the production has already said isn't an element, stays out. If the four-hour deadline arrives first, the result names which departments finished and which still need a look, while the committed breakdown stays in place.
The date and sign-off work now finishes on the item. Change a workflow's need-by and the review shows every date that follows it before anything moves. Pin one step and that promise saves on its own; the other dates stay where they are and come back as a counted list to review. The same page now runs the department's own sign-off from submission through Approve, Request changes, or Reject. If somebody skips a sign-off with a reason, it stands on the Sign-off overrides register until it is signed or the department head or schedule office marks it seen. Changes names who acted, what was there before, and who can put it back.
The Episodes board tells you when nothing is assigned to you yet. On a season that already has episodes, a crew member with no seat no longer reads "No episodes yet" or gets sent to set the season up. It says none are assigned to them, and to ask the production office about that.
On a series, the Script page lists each episode's scenes together. Scene 3 of episode 1 no longer sits next to scene 3 of episode 2. Each episode's run has a heading you can open. A scene that isn't in an episode yet sits under its own heading. A feature film's list is unchanged.
An element that didn't make the list can wait on the tray. If the parse finds something it can't place, you get a card: yes, add it; no, not an element; or ask me later. Yes puts it on the scene as an item for that department, never as a person. No means it won't ask again.
Notes on a version come back to the person who made it. When a director or a department head writes notes, that version shows up on the submitter's Today page, grouped so they can open each one and mark them seen. The notes stay on the version either way.
Folder questions now go to the department head. A wardrobe folder waits on costume, a VFX folder on visual effects — not on a random admin. The person who owns it sees one card on Today and a page of every open question, with pictures they can click and a way to answer a handful at once. Settings still shows the whole list, and names who each one went to.
A department's Workflows tab lists every workflow, and every item without one. Search, sort, and filters live in the address. Tick the rows the filter shows and the count names both units — items and workflows — when they differ. Timeline draws only dates somebody typed: a hollow start tick and a runway only where Start by and due are both there. Type a Start by on a step; it saves on blur, or it tells you it didn't. If that day has passed and the step still hasn't started, the card offers Start it, Move the start, or Mark it seen. The item page can draw the steps as a graph; more than thirty live steps says so instead of guessing a drawing. The old Flow tab is now Turnovers.
You can customise a workflow on the item, or start one from a department's starting point. On a set, a character, a location, or a shot: add a step, duplicate one, reorder, draw a wait, or add a sign-off — review the change, then apply it. Starting points sit under the steps. Every starter says how far this release can take it before you apply, and a banner if it has not been reviewed for this show yet. Apply, merge, replace, or add another workflow. A change can cover fifty workflows at a time.
A shot, a set, a character, and a location now have a page. Open one from the graph and you get its steps — every department that has a workflow on it, side by side, with who signs each step and what's waiting. The trail is on the same page; if a change can come back, Put it back is right there. A shot also shows its plates and versions. An asset still opens on the page it always had, and that page now carries the same steps and trail underneath. Vendors on a shot wait for a later release.
Type S1E5 in search and the episode comes up. On a series, the palette now finds an episode by its code — S1E5, s1e5, 1x5, or the studio's own production code — and opens that episode. A code that isn't on this show is simply no match. Feature films stay as they were.

2026-09-01

Picture descriptions have a daily limit, and HQ sets it. Every picture that lands gets a description and search tags; now each production has a daily allowance for that work, and HQ → Models is where the switch and the amount live. When a show reaches its limit, files still land and file as normal, the descriptions simply wait for the day to roll over, and the activity feed says so once. The tab shows each show's spend today, how many pictures were described, and how many were skipped.
A numbered picture sequence arrives as one item. Drop fifty EXRs, DPXs, or numbered stills in one folder and Throughline shows one row with the frame range and a picture from the middle. Folder questions count it once, the filing plan moves it once, and the looking pass reads that one picture. Every frame is still checked during the move. A missing frame stays part of the same sequence; an unreadable frame leaves the blank stand-in and never holds up the arrival.
A task's folder is laid out like everyone else's. Renders go in renders/, attachments in reference/, note frames in notes/ inside the task's room. Earlier today a task's files sat loose at its root; planning the re-file again puts them where every other folder keeps them.
A photo dropped on the filespace now shows the picture itself. The agent on the mount makes a small JPEG from the file and the card uses that. The original never leaves the drive. A clip uses one frame when the machine can pull one; if it can't, the card stays on the blank stand-in and ingest keeps moving.
A wardrobe task files with the character, not under casting. Link what the task is for — a character, an asset, a location, a scene, or a shot — on the task's own page. Costume work then sits in that person's wardrobe folder; hair and makeup sit in theirs. A prop task stays with the character's main folder. Nothing moves later just because the department on the card changed. Three costume tasks already on the pilot — Mara's slicker, Teddy's uniform, Wren's rehab-tech build — pick up that link the next time the show boots.
Storage setup and teardown finish clean, even when the provider is slow to answer. Setting up a show's drive, syncing folders, checking health and tearing one down no longer step on each other when a request runs long. A teardown the provider refuses stays marked “tearing down” and holds off any new setup until a retry finishes it, instead of failing quietly. If the provider answers late with an account we had already given up on, that account is cleaned up and its key replaced, nothing stale left behind. A drive name stays taken until its cleanup is done, so a new show can't grab a name still being cleared and an old cleanup can't delete the show that took the name after it. Moving an approved file is now crash-safe: nothing gets overwritten, and a file can't end up stranded outside the show's drive if the machine dies mid-move. Each show's storage key is listed in Settings, but Throughline manages it; it can't be typed over or cleared by hand.
Pictures now arrive with words of their own. After an upload, a confirmed proposal, a real filespace preview, or a finished render, Throughline queues a background look. A successful look adds a short plain-English description plus useful lowercase production tags. The description appears on the file card, and a word found only in those tags finds the picture in Canvas and in the jump-to box. A clip uses its sampled frame when one exists. A failed or unreadable look leaves existing words untouched. Shows without a connected visual provider — or with the pass switched off — keep working with empty fields; there is no surprise backfill or failed upload.
A photo dropped on the filespace shows up in the library. Put a still in a character's folder — or a location's, or any other home the show already has on the drive — and it lands on that card. The name comes from the file; a tiny stand-in holds the place until the mounted agent sends its bounded preview. Search finds it by that name and by the description and tags the looking pass writes from that preview. Searching from Canvas or the jump-to box stays inside this show.
A scene, a shot, and the work under them now get folders. A scene that already holds a photo or a render lives under breakdown, named like a person would say it — scene number plus the place. A shot lives under VFX; a real shot code keeps the caps the vendor wrote. A task never gets a folder of its own: it sits under the scene or shot it belongs to, and its files drop straight in. Camera, editorial, and DIT one-offs still stay put until there's a home for them.
The newest model lane has to match the shape we asked for. A parse sent down that lane now carries the same contract every other lane does — extra fields don't sneak through. Cache and thinking-effort controls stay off until we've proven them.
Another model lane is available. HQ → Models can send a task down the new lane the same way it already sends one to the others. That lane bills a flat rate instead of by the call, so a dollar total can't catch a runaway; our own testing counts calls against a ceiling and stops there instead.
The crew can make useful folders without fighting the filing system. Familiar additions such as old, archive, wip, fittings and dated folders are kept where the crew made them. reference and notes are folders Throughline makes itself, so they stay off that list. If Throughline cannot tell what a folder means, Settings asks in plain language and leaves every file in place until an admin answers. “It shouldn’t be here” adds the files to the same reviewable filing plan as every other move — it never moves them immediately. Wardrobe can also use 01_prep/wardrobe/_inbox/: clear character matches join the next plan, while uncertain filenames come back as questions instead of being guessed or forgotten.
A file that comes back after it's been filed shows up on the next plan. If a vendor drops the same path again, or two older files now want the same name, the Filing card says so — a stay with a reason, never a silent skip. The mover also refuses a camera-roll landing that hides behind a folder symlink, and it will not treat a stray file at the destination as "already moved." Its durable move journal now also recovers the exact approved file after a restart if a destination folder was renamed during placement; the journal stays until Throughline receives the ack.
Approving a plan that was already replaced says so. If someone plans again or answers a storage question before the last plan is approved, Approve names the miss instead of quietly approving a changed plan. Two admins can never add unseen files under a plan somebody already reviewed. A file the agent reports twice under the same identity is counted as a repeat, not an error.
A costume, a prop, a location still now gets its own folder. Characters, locations, and shoot days already had homes on the filespace. Anything on the board that already holds a photo or a render gets one too, in that department's prep area — wardrobe and hair items sit beside their character folders.
A reused access-key number no longer wedges the usage count. If an old Throughline access key was deleted and the database reused its id, today's calls stopped counting. That leftover row is cleared, and the counter keeps moving.
There's a page for when a key or a machine looks compromised. Contain the credential first, then look. The runbook lives with the rest of the engineering docs (docs/SECURITY-INCIDENT-RESPONSE.md).

2026-08-31

Existing footage no longer becomes a cast booking just because the script transcribes its dialogue. A named public figure in identifiable news, broadcast or released footage stays a clearance item; a fictional video call or screen performance made for this production stays cast. If the breakdown's own note points both ways, the review tray now asks “clearance or cast?” before the role can silently reach the cast board.
The show's filespace starts organizing itself. Settings gains a Filing section on the storage card: every character, location, shoot day, and item that already holds a photo or a render gets its own folder there, created automatically the moment the person, place, or item is added. A Plan the re-file view lays out where each stray file on the drive should move and why, grouped by reason, with an honest "not sure" pile kept separate from the moves it's confident about — and nothing moves until the admin presses Approve. After that, the card reads back what's queued, what's moved, and what didn't move. Underneath, an agent now watches the filespace directly: it picks up a new file within seconds of it landing, and comes back clean after a crash mid-write instead of missing the file or counting it twice.

2026-08-30

A Day Out of Days lands on the board as its own document. Once the production office approves it, a DOOD PDF reads onto the board one cast line at a time, instead of sitting as a file nobody could see next to the schedule. See Call sheets and day-outs.
Archival footage named in a script stays a clearance item on the breakdown now, instead of showing up as a phantom cast member.

2026-08-29

Crew get their own Day Out of Days. The cast DOOD ships for every production; a crew DOOD is now a per-show switch a production can turn on. Once it's on, a crew member's own held days show on Today, and their contact card gets a booking editor for holding and releasing days. See Call sheets and day-outs.
A professionally numbered shooting script keeps its own scene numbers. When a draft carries a complete set of margin numbers — lettered inserts, OMITTED gaps and all — the breakdown now files each scene's cast and elements against the number printed in the margin instead of its position in the document. An unnumbered spec draft parses exactly as before.
The schedule board tells the truth about dates, and about who's reading it. Every shoot day now carries a real calendar date beside the label the office typed, instead of guessing one that was never stated. Booking terms and open questions on a day stay with the crew office, the budget office, and admins; a crew member reading their own line sees it without the contract attached.
The duplicates review now asks its questions in plain production language, and a straight "no" is taken as an answer.

2026-08-28

Two characters named in the same line no longer merge into one. A joint cue like "AL AND ROSE" used to get claimed by whichever character the pass committed first, silently folding the other person's lines onto their card. Both people now keep their own identity on the board.
Breakdown elements get more departments to land on. Picture Vehicles is now its own department in the registry, Specialty Units is renamed and redescribed for armory and animals, and the element pass can now name virtual-production, greens and specialty-units work that previously had nowhere to file.
Grok joins the model providers a production can connect its own key to. See Your provider keys.
When a model declines to answer a question about a script, the breakdown now tries once more with a more capable model before giving up on the run.

2026-08-27

A shooting script with no sluglines still breaks down. Numbered headings that aren't standard INT./EXT. lines — era stamps, OMITTED markers, plain caps headings — are now read using the script's own margin numbers, instead of going unrecognized by the parser.

2026-08-26

The board stops showing its own workings. Stray PENDING markers next to a role that's already booked are gone, printed page numbers no longer leak into scene text, and a production can renumber its scenes outright — a cold open becomes scene 1, with every lettered scene riding along.
A production can brief the breakdown before it runs. A methodology note and typed casting declarations — "these two Bobs are cast separately," "one performer plays all four, de-aged" — now live in Settings until they're signed off, and travel into every prompt the parse makes. The review tray shows when a brief guided a run.
A scene description now always ends on a full sentence or a clean word, never mid-clause. An exported breakdown is named for the draft it breaks down, not the day someone happened to export it.

2026-08-25

An oversized scene can be split right from the review tray. Accepting a scene-boundary card now splits it out into its own lettered scene (59 becomes 59, 59A, and so on), re-measuring page-eighths and moving cast and elements onto the right sheet.
The review tray's cards ask sharper questions. A twin card folds two names into one on "same person," or books both people on "two different people"; a performer-count card asks how many performers are involved and which scenes each covers; the duplicates panel stops offering a re-scan once a real intelligence run has already answered.
The breakdown sheet no longer prints a raw candidate id where an element name belongs. Approving paperwork while a breakdown review is still open on it now says so, instead of quietly approving something nobody decided.

2026-08-24

Uncertain cast now shows on the board, marked provisional, instead of waiting out of sight in the tray. A principal who reads as ambiguous no longer disappears until someone confirms them by hand — they're on the board with a provisional badge, and the review tray catches up around them.
The review tray now batches the easy calls: confirmations that don't need a second look sit together under "Needs your eye," and Confirm all clears the whole batch in one action with a per-row undo.

2026-08-22

The breakdown reads a script the way a line producer does. A model now judges every candidate character and element with the whole story in view, resolves who's who as casting roles, and sends anything it isn't sure of to a human review tray — rather than guessing.
The script drop zone now warns, in red, when no model is connected — parsing degrades a lot without one, and it says so before a script goes in, not after.

2026-08-21

Uploading a script no longer freezes the page. Ingest now runs as a background job with visible progress, instead of a request that could run for minutes and time out the upload.
Character names that are really the same person now fold together. An inline aside like "DACE ON THE PHONE" or an occupational title like "CO-PILOT DACE" now folds onto the character it's describing instead of minting a second card — and it refuses to guess when a fold could be wrong.
Every item in a one-liner's Cast or category cell gets its own line inside the cell now, instead of a comma list that wraps mid-word. The script page names the model that will actually read the pages, instead of a generic nudge to go connect one. The Pulse send log moved from Settings to the Jobs page, where "what did Pulse do" already lives.

2026-08-20

The breakdown leaves as paperwork. A one-line schedule as XLSX — a script-order sheet always, a shooting-order sheet whenever the show has shoot days — or the same one-liner as CSV, or the long-format element interchange as CSV or JSON. One door: GET /api/breakdown/export?fmt=xlsx|csv|elements_csv|elements_json, signed-in and clamped to what the caller may already see. Cast carries board numbers assigned by first appearance the way every AD one-liner already does, the header block names the revision the pages actually carry, and an episodic export carries its episode in the rows and in the filename so two episodes exported the same day don't collide.
Scenes carry their own length, in eighths of a page. A scene reads 1 4/8, not a rounded page count, and every draft before the shooting script locks is a dated White page instead of a color collision. A length read from anything but a PDF's own page breaks says so — it's marked estimated rather than presented as measured.
A crew member meets Sign in with Slack on their own first morning. Connecting a Slack account is something a person does for themselves, from their contact card or the offer waiting on Today — not something they have to go find. When Slack meets an account it doesn't recognize, it now points at that one-click fix before falling back to the production office. See Slack.
The assistant answers from the same graph the screen is drawing. It matches a scene number with or without leading zeros, groups a character's scenes by episode without dropping any past the cap, and names a node's other connections instead of going quiet. It no longer calls data absent when it never looked for it.
Settings says who each control affects. Everything that shapes the whole production — message routing, Pulse, reading access, department appearance, channels — sits in one Production section, closed by default. A person's own controls, including their own connected apps, stay in their own column.

2026-08-19

Slack moved to where each half belongs. Everything that is a fact about the SHOW — install, mode, health, disconnect, the simulator and the coordinator queue — is now one panel in Settings → Slack, and Slack's own consent screen returns you there, so the handshake begins and ends on one screen. Which Slack account belongs to a person is a fact about that PERSON, so it moved onto their contact card in Crew, beside their email, phone and time zone. The separate Channels screen is gone; every API route it used stays exactly where it was. See Slack.
Slack connects in one click, and leaving really revokes. The install starts in Settings, goes to Slack's own consent screen, and comes back to Settings. Disconnecting by any door — from here, from Slack's admin screen, or by uninstalling the app — drops the tokens for good. See Installing the app.
One question gets one answer, in the thread it was asked in. The events door acknowledges Slack immediately, remembers what it's already seen so a retry doesn't answer twice, and replies in the thread rather than the channel.
api.criticalperiod.co and mcp.criticalperiod.co are the addresses. Both were provisional in these docs until now. They're not — build against them.

2026-08-18

Throughline is its own authorization server. An MCP-aware client can now connect with nothing but a URL: protected-resource and authorization-server metadata, dynamic client registration, PKCE, and a consent screen you sign in to as yourself, so what the connection reaches is exactly what you reach. Refresh tokens rotate, a replayed one is detected and kills the grant, POST /oauth/revoke does what RFC 7009 says, and deactivating a person reaches their tokens rather than waiting for them to expire. Access keys are unchanged — this is a third way to hold a credential, not a replacement. See Discovery.
Connected apps, in Settings. Every grant standing on the show, who holds it and what it can reach, and a revoke that bites on the next call.
The audit trail answers every question it promised. A complete event taxonomy, 400 days of memory, and alerts that reach a person instead of a log file.
A failed ingest says so at the drop zone, where the file was dropped, rather than only in a banner at the top of the page.

2026-08-12

Assignments go down to the scene, and sideways as a grant. A person can be assigned particular scenes rather than a whole episode, and a may-see grant can widen what somebody reaches. A grant only ever opens a door — it can never close one the show already opened — and the same width governs script, media, and Slack channel membership alike, so those three can't drift apart.
The episode page tells the truth about who's on it, and someone granted a read in from outside leaves notes that carry their name.

2026-08-11

A day's paper opens where the day is, and leaves as one zip. Call sheets, sides and maps are readable on the day itself instead of downloading one at a time, and the whole day's packet comes down together. See Call sheets and day-outs.
Episode budgets get a door, and a series' board and spend reports backfill themselves rather than starting empty.
Department appearance collapses to one default and a list of exceptions, instead of a setting per department. See Per-department display names.
WhatsApp inbound reaches the pilot. Twilio's own request signature is the gate, so a crew member's reply arrives without a shared secret in the URL.
The docs learn to be read by machines as well as people. Every page is served as raw Markdown beside its HTML, and the site publishes llms.txt, llms-full.txt, sitemap.xml, and the API's OpenAPI document.

2026-08-10

MCP server. POST /mcp — the same REST v1 surface over Streamable HTTP JSON-RPC, authenticated on the same access keys, with zero authorization logic of its own: every tool and resource calls the identical function its matching REST endpoint calls, so a key gets the same clamped answer and the same refusal on either door. See MCP.
Context bundles, turnovers, media, and a vendor money read. /v1 grows a second shape of endpoint — one call that assembles everything about a scene or a character, a vendor's own turnovers and submissions, media read by id, and a vendor's own purchase orders — all clamped to the calling key. See One round trip, clamped to the caller.
Engagement-bound access keys. A Throughline access key can now be minted against one vendor organization's engagement instead of the whole production — a vendor's own automation, reaching exactly what a person seated on that engagement would reach. See Keys.
Slack. Install a workspace for a production, bind crew accounts, and let the show provision its own channels off the crew list. Notifications route by urgency — interrupt, ambient, digest — and an approval's whole lifecycle lands on one message, updated in place, rather than a fresh post every time. See Slack.
Forge builds behind a kernel jail for vendor and API access. Building or running generated code through a vendor's own seat or an access key now requires the deploy's kernel-level sandbox where the host provides one; the production's own crew is unaffected.
Public docs launched (pilot preview).

2026-08-09

Access key metering. Every authenticated /v1 request is now counted per key, per day — visible next to the key in Settings as a request count and a last-used timestamp. Nothing is gated or billed on it yet; access stays seat-priced. See Keys.
Two new scopes. submit:turnover and read:money join read and write — see Scopes.
Episodic productions. Seasons and episodes, per-episode budgets and schedule filtering, now reflected across the read surface: an episodic show's scenes and schedule strips carry episode_id/episode_code, and /schedule and /script-changes both take an episode_id filter.