Decoupage Alpha Testing Protocol

This walks you through specific things to try, in roughly the order a real workflow would touch them. The point is to make sure you hit the features that matter.

(The app used to be called Fathom. It's Decoupage Screenwriting now — same app, new name. If you saw a Fathom build before, this is it.)

You don't have to do these in order, or do all of them. But the more you cover, the more useful your feedback. Each section runs 5–30 minutes depending on how deep you go.

Sections are marked essential, important, or optional. If you've only got an hour, do the essentials. I'd rather have thorough feedback on four sections than rushed coverage of all of them — reach a stopping point, decide you're done, send back what you have.

Make notes as you go — what worked, what didn't, what felt awkward, what surprised you. Use the Findings Log I sent with this one.

Section 1: First Launch — Sample + Quickstart (optional, 5 minutes)

Goal: see how Decoupage greets a brand-new user. Best on a true first launch, but you can try the pieces any time.

1. On first launch, the Welcome window appears. Note whether the Reference panel opens on its own to a "Quickstart" tab (it's meant to auto-open exactly once, on the very first launch — if you've already opened the app a few times, you may have missed it; you can always reach it via ⌥⌘R).
2. In the Welcome window, click "Explore a sample." A small pre-filled sample screenplay opens — it's a synthetic example (not a real script) with a bit of cast, breakdown, schedule, and call-sheet data already in place. It opens as a fresh untitled copy, so poke around freely; you won't damage anything.
3. Open the Reference panel (⌥⌘R) and look at the Quickstart tab — a one-screen tour of the app's tabs and features. Try clicking a feature row and confirm it jumps you to that tab.

Note any of these:

• Did the first-launch greeting (the auto-opened Quickstart) feel helpful, or in the way?
• Did "Explore a sample" give you a useful sense of what the app does without having to write a script first?
• Was the Quickstart tour clear? Did the click-to-jump rows land where you expected?
• Anything you wished the sample or the tour had shown you?

Section 2: First Document (essential, 15 minutes)

Goal: get a feel for the writing experience.

1. Open Decoupage Screenwriting. The Welcome window should appear.
2. Click "Start a new screenplay." A blank document opens.
3. Switch to the Title Page tab (top of the document).
4. Click the pencil icon to edit the title page. Fill in:
   • Title
   • Credit
   • Author (your name)
   • Draft date
5. Click Update Page. Switch to the Write tab.
6. Type a complete scene from scratch. Include:
   • A scene heading (INT. or EXT.)
   • At least two action paragraphs
   • At least two characters in dialogue
   • At least one parenthetical
   • At least one transition (e.g., CUT TO:)

Note any of these:

• Did the text format correctly as you typed? (Scene heading recognized, character cues centered, dialogue indented, etc.)
• Did anything feel wrong, slow, or surprising?
• Did the cursor move where you expected?
• Did keyboard shortcuts feel intuitive?

Section 3: Formatting (important, 10 minutes)

Goal: try the Format menu and inline markup.

1. In the scene you just wrote, select a word in some action text.
2. Press ⌘B. The word should become bold.
3. Click elsewhere. The asterisks around the word should disappear, leaving just visual bold.
4. Click on the bolded word. The asterisks should reveal.
5. Try the same with ⌘I (italic) and ⌘U (underline).
6. Open the Format menu. Try each menu item with selected text:
   • Center
   • Insert Note
   • Insert Boneyard
   • Insert Link
   • Insert Synopsis
   • Insert Section
   • Insert Page Break

Note any of these:

• Does the toggle behavior feel right? (Pressing ⌘B on already-bold text should un-bold it.)
• Did any combination produce weird results (e.g., overlapping markers)?
• Did the Format menu commands feel discoverable?
• Did the keyboard shortcuts feel natural?

Section 4: Notes and Boneyard (important, 10 minutes)

Goal: try writer notes and commented-out content.

1. Add a note somewhere in your scene: [[Need to revise this dialogue]]
2. Click outside the note. It should appear with a yellow tint.
3. Switch to the sidebar and find the Scenes tab. Your note should appear under the scene heading.
4. Click the note in the sidebar. The editor should scroll to that note.
5. Add some boneyard: /* This was an earlier version */
6. Click outside. The boneyard should appear grey with strikethrough.
7. Confirm the boneyard does NOT appear in the sidebar navigator.

Note any of these:

• Does the visual treatment make notes vs. boneyard immediately distinguishable?
• Is the sidebar navigation useful?
• Did you find yourself wanting to do something the navigator doesn't let you do?
• Anything that felt slow or laggy?

Section 5: Preview Tab (essential, 10 minutes)

Goal: see your script the way a reader will.

1. With your scene from Section 2 still open, switch to the Preview tab.
2. You should see your script laid out on a virtual page — title page first (if you filled in title-page metadata), then the body.
3. Scroll through. Confirm scene headings, action, character cues, dialogue, parentheticals, transitions all sit where you'd expect on a screenplay page.
4. At the top of the Preview tab, find the Production View / Writer's View toggle.
5. Try Writer's View. Any notes ([[like this]]) should appear inline with a yellow tint — this is what you see while editing.
6. Switch back to Production View. The notes should disappear — this is what the script looks like when handed to anyone other than you.
7. Click a scene in the sidebar navigator. Preview should scroll to that scene.

Note any of these:

• Did the page layout look like a real screenplay page?
• Was the Production / Writer's toggle clear about what it does?
• Did the writer's-view note rendering match what you saw in the editor?
• Did the scroll-to-scene navigation feel responsive?
• Did anything look misaligned, clipped, or surprising?

Section 6: Cast and Breakdown (important, 15 minutes)

Goal: try the production-metadata views. Cast is fully wired. Its exports come off its own Export menu — a Cast List sheet (PDF / Markdown / CSV) and the two Casting Breakdown PDFs (that PDF gets its own walkthrough in Section 7; the wider report family is Section 9). Breakdown / Stripboard are browse-and-correct surfaces for now.

1. Switch to the Cast tab. You should see the characters from your scene listed.
2. Click on a character. Fill in some metadata:
   • Performer — the actor / reader you've cast (or left blank if not yet assigned)
   • Tier (e.g., Lead, Day Player, etc.)
   • Age range
   • Gender
   • Physical description
   • Character description
   • Casting notes
   • Union status
3. Notice the profile uses progressive disclosure — Casting Category and Demographics stay open, while Description, Notes to Casting, and Appears-in are collapsible (Appears-in starts collapsed). Expand/collapse a couple and note whether it makes the form easier to scan or just adds clicks.
4. Gender suggestion (every Mac). Leave a character's Gender unset for a moment. If the script gives a signal — an honorific like "Mr." near the name, or a lean in the gendered words across their scenes — Decoupage shows a lightbulb suggestion with an evidence line and an Apply button. It's a suggestion you confirm, never auto-applied. (If the script gives no signal and your Mac has Apple Intelligence, an "Infer with AI" button appears instead, badged AI — on-device. On a Mac without Apple Intelligence there's no AI button; that's expected.) Note whether the suggestion (and its evidence) felt trustworthy, and whether Apply did what you expected.
5. If your script has a character who appears under multiple names (alters, a dual-language character speaking under two cue variants), pick the alter and use the Alias of field to point at the canonical character. The two will collapse into one entry in the Casting Breakdown PDF (Section 7) and the canonical's scene + page counts will combine across both.
6. In the Cast tab's toolbar, open the Export menu and choose Cast List… (PDF / Markdown / CSV). This opens an export sheet where you pick the format (PDF / Markdown / CSV, plus DOCX in the grid layout), the scope, and which columns to include. Try a Markdown and a CSV export, save each to your Desktop, and open them. The two Casting Breakdown PDFs are in this same Export menu (Section 7 covers them), and the whole report family is also reachable from the top-level Reports menu (Section 9).
7. Switch to the Breakdown tab. You should see your scene with elements detected across the 14 categories (cast, props, wardrobe, SFX, etc.).
8. Try correcting the auto-detected breakdown for one scene:
   • Add a prop
   • Mark a special effect
   • Add wardrobe notes Overrides persist with the document. The Breakdown editing surface is still review-and-correct — but its reports now export (Scene Breakdown Sheet PDF, Scene Breakdown CSV, Department Pull Lists, the Background/Extras reports), all under the Reports → Breakdown menu. Section 9 walks through them.
9. Inside the Breakdown row, try the Shots sub-section. You can add per-scene shot rows (slug, description). These feed the Shot List export (Reports → Camera → Shot List PDF/CSV — Section 9).
10. Switch to the Stripboard tab. Your scenes should appear as colored strips, ordered as they sit in the script, with page eighths and cast counts. The Stripboard is a read-only visual in this build — you can filter and inspect, but you can't reorder strips into shoot days, group by day, or export. Scheduling now lives in the dedicated Schedule tab (Section 8).
11. Switch to the Department Notes tab. Pick a department, type some notes. The 13 departments are independent free-text notebooks tied to the document.

Note any of these:

• Did the Cast view feel like a useful tool for casting prep?
• Did the new Performer field land where you'd expect it on a profile?
• Did the progressive-disclosure form (open Demographics, collapsed Appears-in) feel cleaner, or fiddly?
• Tab between fields (in a cast profile, Document Settings, Day details, a Supporting Materials caption) should now advance focus to the next field rather than getting stuck. If Tab traps anywhere instead of moving on, that's worth a note — it's a fix this round.
• If you saw a gender suggestion: did the evidence make sense, and did Apply feel safe? (And if you're on an Apple-Intelligence Mac and tried "Infer with AI," how did that land?)
• If you tried Alias of: did the link feel intuitive, or weird?
• Did the Cast Markdown / CSV exports look like something you'd actually share with a casting director?
• Did the auto-detected breakdown catch the elements you expected, or miss things / catch wrong things?
• Was the Stripboard's read-only-visual framing clear, or did you reach for scheduling actions that don't exist yet?
• Did Department Notes feel like something you'd actually use?
• Anything missing from these views that you'd want?

Section 7: Casting Breakdown PDF (essential, 10 minutes)

Goal: generate the casting-side production document and confirm it looks like something you'd send.

The Casting Breakdown PDF is the first production-document export. Two variants:

• Submission — descriptive role-record fields only, no scene/page counts, grouped by tier. This is what goes to casting.
• Production — same fields plus per-character scene count, page count, and performer assignment. The internal planning version.

1. Make sure you've filled in at least two character profiles in the Cast tab (Section 6). If your script also has a character you marked as an Alias-of another, even better.
2. Reports → Casting → Casting Breakdown — Submission (PDF)… (also in the Cast tab's Export menu). Decoupage shows a preview; Export it, save to your Desktop, and open the result in Preview.app.
3. The PDF should have a title (the screenplay title), a per-tier grouping with tier headers, and one block per character. Each block: bolded name, then a one-line rubric like "Lead · Woman · 30–40 · Latina", then the description, then casting notes / union if present. No scene counts, no page counts, no performer name — the submission variant deliberately suppresses those.
4. Reports → Casting → Casting Breakdown — Production (PDF)… Preview, Export, save and open. Same layout, but each block now includes scene count, page count, and the performer name when set.
5. If you set up an Alias-of relationship in Cast (step 3 of Section 6): in the Casting Breakdown PDF, the alias should NOT appear as its own block — instead, its description / notes should fold in as a nested sub-section under the canonical character. In the production variant, the canonical's scene + page counts should reflect the combined (deduplicated) totals across canonical + alias.
6. Quick sanity check: Fountain inline markup like _underline_ or *italic* in the script title or descriptions should appear as PLAIN text in the PDF — the casting breakdown is a plain-text production document, not a styled-text one.

Note any of these:

• Did the Submission variant feel like a real casting submission?
• Did the Production variant feel like something useful for internal planning?
• If you used Alias-of: did the collapse + fold-in read sensibly, or weird?
• Did anything get clipped, run off the page, or look misaligned?
• Was the difference between the two variants obvious from the output (rather than just from the filename)?

Section 8: Schedule Tab (important, 15 minutes)

Goal: try the new scheduling primitive — auto-grouping plus manual day adjustment.

The Schedule tab is new this round. It's the data foundation for day-out-of-days, full shot-list exports, and the production-budget aggregate — none of those are built yet. Schedule is the under-layer.

1. Switch to the Schedule tab for the first time on a document. Decoupage should auto-group your scenes into shoot days by (location + INT/EXT + day/night) — i.e. scenes that share a location and the same time-of-day classification land on the same default day.
2. Look at the list. Each day shows: a day label (Day 1, Day 2…), the location-INT/EXT-time-of-day grouping that defined it, the scenes assigned to that day, and a per-day page total.
3. Try the manual adjustments:
   • Drag a scene from one day to another. The page totals should update.
   • Drag a scene to "Unscheduled" if that's an option in the UI — pulls a scene out of any day.
   • Rename / edit a day — open a day's "Day details…" (it replaced the old "Rename…"). Beyond the label you can set the day's date, call time, shoot address, and notes — these feed the Call Sheet (step 6).
   • Move a day earlier / later in the sequence.
   • Merge two adjacent days into one.
   • Split a day at a scene boundary.
4. Save the document, close it, reopen. The schedule should be preserved exactly as you left it (including anything you typed into "Day details…").
5. Try Reports → Scheduling → Master Schedule (PDF)… — generates a production-planning master document combining schedule + cast + breakdown. Decoupage shows you a preview first; click Export to save it, then open it. (The full report family gets its own walkthrough in Section 9 below.)
6. Call Sheets (new). First, set a couple of header fields: open Document Settings and fill in the production company and director; then open one day's "Day details…" and set its date, call time, and shoot address. Now choose Reports → Scheduling → Call Sheets (PDF)…. Decoupage previews the call sheet; Export it, save to your Desktop, and open it. You should see a per-shoot-day sheet: the day's scenes, cast grouped by tier, locations, page totals, and the header fields you set. (Fields you leave blank just come out blank — it won't invent a call time.)

Note any of these:

• Did the auto-grouping land somewhere sensible for your script, or did you immediately want to override it?
• Was each manual operation (drag, Day details, merge, split) discoverable, or did you have to hunt?
• Did anything fail to persist when you closed and reopened?
• Did the Master Schedule export read like a useful planning document, or did it need more / less detail?
• Did the Call Sheet read like something you'd actually hand a crew? Did the header fields (company / director / date / call time / address) show up where you'd expect, and did the scenes / cast / locations for the day look right?
• This is the first version of scheduling in Decoupage — what did you expect to be there that wasn't?

Section 9: Production Reports — the Reports menu (important, 15 minutes)

Goal: exercise the production-document exports and confirm each one opens, looks right, and contains the scenes / characters you'd expect.

There's a top-level Reports menu in the menu bar. It collects the production documents, grouped by department (Casting, Scheduling, Locations, Camera, Breakdown), plus Coverage and Structure X-ray. Everything here is deterministic — built from data already in your document — so it runs on every Mac, no Apple Intelligence needed. (Coverage is the one AI item in that menu; it gets its own walkthrough in Section 19.)

Two things to know first:

• For the PDF and CSV reports, Decoupage shows you a preview first. Look it over in a sheet, then click Export to write the file. Save to your Desktop and open it.
• A report is only as full as the data behind it. These read your Cast profiles, Breakdown tags, Schedule days, and shot rows. Enter little, get sparse output — and the report should say it's sparse (e.g. "No shots planned yet") rather than invent detail. That honesty is worth checking too. For the richest test, do Sections 6–8 first so there's real data to report on.

Run as many of these as you have appetite for — you don't have to do all of them. For each: does the preview appear? Does the file open? Are the right scenes / characters present? Does it look like a document you'd hand someone?

1. Reports → Casting → Cast List… — opens an export sheet. Pick a format (PDF / Markdown / CSV; DOCX appears in the grid layout), a scope (whole script / a single character / a scene range), and which columns to include (there's a "We Make Movies" preset). Export and open. Confirm the characters you filled in are present with the columns you chose. (This same sheet is also reachable from the Cast tab's own Export menu — Section 6.)
2. Reports → Casting → Casting Submission — Posting Form (PDF)… and Casting — Creative (PDF)… — two more casting-facing documents. Export each, open, and note whether they read like something you'd send. (The Casting Breakdown — Submission / Production PDFs also live here; they got their full walkthrough in Section 7, so no need to repeat.)
3. Reports → Scheduling → Day-out-of-Days (CSV)… — the cast-availability grid (which performer works which day). Open in Numbers / Excel; confirm it's a sensible grid. (Reports → Scheduling also has Call Sheets (PDF)… — the per-shoot-day call sheet; it got its full walkthrough in Section 8 step 6, so no need to repeat here.)
4. Reports → Locations → Location List (PDF)… and (CSV)… — every location with the scenes shot there. Confirm your locations are present and the scene assignments look right.
5. Reports → Camera → Shot List (PDF)… and (CSV)… — your planned shots per scene. If you added shot rows in the Breakdown tab's Shots sub-section (Section 6), they should appear; scenes with no shots should surface as gaps ("No shots planned yet"), not be silently dropped.
6. Reports → Breakdown — try Scene Breakdown Sheet (PDF)… / Scene Breakdown (CSV)…, Department Pull Lists (PDF/CSV)…, and the Background / Extras Report (PDF)…. These read your 14-category breakdown tags. Background Breakdown… opens a short guided sheet (it scans for crowd nouns; you confirm/edit the buckets) before producing its PDF.
7. Reports → Music Cues… — a master-detail entry sheet. The left rail lists song/cue candidates a deterministic scan found in your action lines; Confirm the real ones, add clearance notes, then Export (PDF / CSV). If you're on a Mac with Apple Intelligence, you'll also see a "Find lyrics with AI" control — it surfaces sung lyrics it spots, badged AI; treat those as suggestions to verify, not gospel. (No AI Mac? The control is hidden and the deterministic detection still works.) Note whether the detected candidates were sensible, whether Confirm + notes + Export felt clear, and — if you tried it — whether the AI lyric finds were useful or noisy.
8. Reports → Production Flags… — a master-detail entry sheet for specialty/risk elements (firearms, stunts, fire, minors, period, …). Deterministic, every Mac — no AI. The candidates carry High / Med / Low burden chips and sort high-burden first; Confirm the real ones and add handling notes. It exports two documents, each PDF or CSV: a Stunt / VFX Breakdown and a Risk Register. Try exporting each. Note whether the flags it caught matched what your script actually has, whether the burden weighting felt right, and whether the two reports read like documents you'd hand a line producer / 1st AD.
9. Reports → Structure X-ray… — this one's an interactive panel, not a file. It overlays the standard three-act template on your script's actual page geometry — where each canonical beat (inciting incident, act breaks, midpoint, climax) lands and the real scene there. Note: it's honest that it's placing the convention on the page math, not claiming to detect your actual midpoint. Look it over; nothing to export.

Note any of these:

• Did each report you tried open cleanly? Anything clipped, run off the page, or misaligned?
• Were the right scenes / characters / locations present — did anything get dropped or double-counted?
• When a report was sparse because you hadn't entered the data, was that clear and honest, or did it look broken?
• Did the preview-then-Export flow feel right, or did you expect a straight save?
• Which of these would you actually use? Which felt like noise?
• Did the Structure X-ray's "we're placing the convention, not detecting it" framing come through, or did it read like a claim about your script?

Section 10: File Handling (important, 10 minutes)

Goal: make sure file operations work properly.

1. Save your document. File → Save (or ⌘S). Pick a location and a filename. New documents save as a .dcpg file (the Decoupage document format). (If you happen to have an old .fathom file from an earlier build, open it and Save — it should open fine and save forward as .dcpg. Worth a quick check if you have one.)
2. Close the document (⌘W or red close button).
3. Reopen it from the Open Recent menu (File → Open Recent).
4. Verify everything is preserved: title page, script body, cast, breakdown, notes.
5. Try File → Duplicate (or ⌘⇧S for Save As). Save a copy.
6. Try File → Export → Fountain... Save the export somewhere. There's an "Include notes in output" checkbox in the save panel — try it both ways and note what changes in the exported .fountain file.
7. Open the exported .fountain file in TextEdit (or another text editor) to verify it's readable.
8. PDF export gets its own section next — for now, just confirm the menu item exists at File → Export → PDF…

Note any of these:

• Was the save process intuitive?
• Did anything get lost on close-and-reopen?
• Was the export readable as plain text?
• Did the "Include notes" toggle behave as you'd expect?
• Did any of these operations feel slow?

Section 11: PDF Export (essential, 15 minutes)

Goal: generate a real screenplay PDF and look it over.

1. With your document open (one with a title page filled in), choose File → Export → PDF...
2. The save panel should appear with a checkbox at the bottom: "Include notes in output." Leave it unchecked for now.
3. Pick a location and save.
4. Open the resulting .pdf in Preview.app (or any PDF reader).
5. Look at the title page. The title should sit roughly at the upper third, centred. Author and credit centred in the middle. Contact bottom-left, draft date bottom-right.
6. Flip to the second page (first body page). There should be no page number in the top-right — that's the convention.
7. Flip to the third page. There should be a page number 2. in the top-right.
8. Scroll through the rest. Scene headings, action, dialogue should all sit where they did in the Preview tab.
9. Try Tools → Scene Numbers → Add Scene Numbers (a later section covers this in more depth — for now, just run it once).
10. Re-export the PDF. Now scene numbers should appear in both margins of the body pages, beside each scene heading.
11. As one last check: re-export with the "Include notes in output" checkbox checked. Any [[notes]] you have should now appear in the PDF body.
12. Revision marks (optional, needs a draft history). In the export panel there's a checkbox: "Mark revisions (changes since last draft)." This only does something if you have a prior draft to compare against, so: make sure you've edited the document at least a little since a saved draft exists (the Drafts timeline — Section 18 — keeps these automatically). Then export with the box on. Open the PDF: lines that changed since your most recent draft snapshot should carry a right-margin asterisk (*), the standard "this line was revised" mark. Export again with the box off — the asterisks should be gone and the PDF otherwise identical. (This is marks-only for now — no coloured revision pages or A-pages yet.)

Note any of these:

• Did the PDF look like a real screenplay?
• Were the title page positions reasonable? (Compared to what you've seen from Highland Pro / Final Draft / sample scripts.)
• Did page numbers appear on the right pages?
• Did scene numbers in the margins look right?
• Did the "Include notes" option do what you expected?
• If you tried revision marks: did the asterisks land beside the lines you'd actually changed? Did "off" produce a clean, unmarked PDF?
• Did the PDF export feel fast or slow?

Section 12: Collaboration — notes on a draft (new this round, 15 minutes)

Goal: run the whole feedback loop — export a Review PDF, fill it in, import the notes back, work with them in the Comments panel. This is new this round and going to testers for the first time, so treat it as an early look, not a polished feature — expect rough edges, and tell me what breaks or what's confusing. Everyone can test it: no cloud, no account, and the "reviewer" can be you.

The idea: you hand a reader a fillable PDF of your script. They write notes in it — any PDF viewer, or printed and penned — and Decoupage reads those notes back in, attached to the right scenes. Don't exercise it exhaustively. A single round-trip and your honest first impression is what I'm after.

1. With a script open (the Section 2 scene, or the sample from Section 1), choose File → Export → "Review PDF (for notes)…" and save it to your Desktop.
2. Open that PDF in Preview.app. You should see your script in a column with a comment box in the right margin beside each scene, and an Overall section at the end (a 1–5 "how it lands" scale plus a few prompts). Type a note or two into a couple of the per-scene boxes, fill in a reviewer name, and answer some of the Overall prompts. Save the PDF. (Optional: print a page and confirm the rail + Overall are usable with a pen.)
3. Back in Decoupage, choose File → Import Review Notes… and pick the PDF you just filled. The notes should come in, and the Comments panel should reveal itself on the right.
4. In the Comments panel: confirm your notes are grouped by the scene they belonged to (plus an Overall/general group). Click a scene header — the editor should jump to that scene. Click a comment's circle to mark it resolved — it should dim.
5. Hit the "+" in the panel to add a comment by hand — type a body, set a reviewer, and pick the scene it attaches to. (This is how you'd transcribe a pen reviewer's notes, or jot your own.)
6. Toggle the Comments panel off and on from the toolbar to confirm it shows/hides cleanly.

(One cosmetic thing to expect, not report: on the exported Review PDF, a scene's right-margin scene number can sit right next to the notes-rail label. Harmless — the boxes still fill and import fine.)

Note any of these:

• Did the Review PDF read like something you'd actually send a reader? Were the per-scene boxes obvious, and big enough?
• Did filling it in Preview (or on paper) feel natural?
• On import, did the notes land on the right scenes? Did anything land in the wrong place or get dropped?
• Did the Comments panel's jump / resolve / manual-add all behave as you'd expect?
• Would you actually use this to get notes on a draft? What's missing for that to be true?

Section 13: Highland Compatibility (optional, 10 minutes)

Goal: if you have any Highland Pro .highland files, try importing one.

1. File → Import Screenplay
2. Select a .highland file
3. Verify Decoupage imports it correctly:
   • Script body is preserved
   • Title page is preserved
   • Characters detected in Cast view
4. If the Highland file already had scene numbers (the #N# syntax, or numbers Highland was managing): they should come through transparently. Look at the Preview tab — scene numbers should appear in the margins if the source had them.
5. Try to edit the script body. You should see a banner saying "Linked to Highland source. Edit in Highland Pro."
6. Try to edit Cast or Breakdown — these should work normally even though the script is read-only.

Note any of these:

• Does the "read-only Highland" workflow make sense?
• Or is it confusing/annoying?
• Did anything get lost in the import?
• If your Highland file had scene numbers — did they come through correctly?

Section 14: Final Draft Compatibility (important, 15 minutes)

Goal: import a real .fdx and confirm Decoupage reads it cleanly. If you have a Final Draft script, run this section. FDX import is the path I most want tested this round — don't treat it as "only if you happen to have one handy" the way the Highland section above is framed. Even a single .fdx open-and-look-around is useful.

1. File → Import Screenplay
2. Select a .fdx file
3. Verify the import:
   • Script body comes through
   • Title page metadata (title, author, credit, draft date) populates the Title Page tab
   • Characters appear in the Cast tab
4. Switch to the Preview tab. If your script has any dual (simultaneous) dialogue — two characters speaking at the same time — they should render side-by-side in two columns, not stacked. If your script has none, this step is a no-op; skip it.
5. Look at the top of the Write tab. There should be a banner: "Linked to Final Draft source. Edit in Final Draft."
6. Try to edit the script body. Edits won't take — the banner explains why. Linked files are read-only in Decoupage; the source of truth stays in Final Draft.
7. Confirm Cast / Breakdown / Stripboard still work normally. The script body is locked, but the production-metadata views remain editable on top of the linked script.

A formatting-marks quirk to expect, not report:

• For imported Final Draft files, the Write tab shows raw Fountain formatting marks — a leading . on scene headings, a ^ on the second line of simultaneous dialogue. That's by design for linked docs (Write shows the underlying source text). Preview and PDF render clean without the marks.

Note any of these:

• Did anything get lost in the import? Specific scenes, character cues, title-page fields, formatting choices?
• If your script has dual dialogue: did it render side-by-side correctly in Preview?
• Does the read-only-linked workflow feel sensible, or weird? (Genuine question — I don't know yet whether this lands for writers who'd actually use it.)
• Anything else worth flagging about how your real Final Draft script lands in Decoupage?

Section 15: Tools → Scene Numbers (optional, 10 minutes)

Goal: try the scene-number bulk commands.

Most useful if you write or manage scripts where scene numbers matter — production-bound work, locked pages. Early-draft writers can probably skip it.

1. Open a script that has no scene numbers yet (the script you wrote in Section 2 is fine).
2. Look at the script text in the Write tab — no #N# markers anywhere.
3. Tools → Scene Numbers → Add Scene Numbers
4. The script text should now show #1#, #2#, etc. at the end of each scene heading line.
5. Check the Preview tab — scene numbers should now appear in both margins beside each scene heading.
6. Try ⌘Z (undo). The numbers should disappear from both the script text and the Preview margins.
7. Re-run Add Scene Numbers. Then try Tools → Scene Numbers → Remove Scene Numbers. The #N# markers should be stripped from the script and the Preview margins should be empty again.
8. If you're feeling adventurous: manually type a custom number like #5A# on one scene, then run Add Scene Numbers on the rest. Your #5A# should be preserved while the unnumbered scenes get sequential numbers above 5.

Note any of these:

• Did the bulk commands do what you expected?
• Was undo / redo behaviour normal?
• Were the margin scene numbers in the PDF and Preview positioned reasonably?
• Did anything feel surprising or counter-intuitive?

Section 16: Read Aloud (important, 10 minutes)

Goal: hear the script in your ear, in two modes.

Read Aloud gives each character their own distinct voice — it plays back like a radio read-through, not one flat narrator. It can use your Mac's installed system (Apple) voices, which work immediately, and Decoupage auto-assigns those across your cast, preferring the highest-quality installed voice for your system language (premium > enhanced > default). Install premium or enhanced voices via System Settings → Accessibility → Spoken Content → System Voice → Manage Voices and Decoupage draws on those instead of the default Samantha. So install a couple of good ones before this section if you can.

Decoupage also has its own higher-quality local Kokoro voices. These download the first time you use them rather than ship in the app — so the first time you trigger Read Aloud (or audio export) with a Kokoro voice, you'll see a progress banner while they download. Once that one-time download finishes, they're ready. If they aren't there the moment you install, that's expected — they arrive on first use. (This first-run download is new and going to other machines for the first time. If it stalls, fails, or the banner behaves oddly, tell me.)

1. With a script open, choose Tools → Read Aloud → Read Entire Script (or whatever the menu item is labeled). The script should start playing from the top.
2. Pause / Resume / Stop — these should be available in the same menu (or via keyboard shortcut) while playback is in flight.
3. Click into the editor on some scene in the middle of the script. Choose Tools → Read Aloud → Read From Current Scene. Playback should start at the scene containing your cursor — not from the top.
4. Listen for a minute. You should hear different voices for different characters, and character cues are NOT read aloud (you won't hear "SARAH" before each of Sarah's lines — the change of voice is what tells speakers apart).
5. Optionally, pin a voice to a specific character: go to the Cast tab, open a character's profile, and use the voice picker (set it to a specific voice, or leave it on "Auto" to let Decoupage choose). If the character has a gender set and the picker is on Auto, Auto is now gender-aware — it shows you the voice it chose, e.g. "Auto — Female voice (Samantha)" — and picks a system voice matching that gender. Then start Read Aloud again and confirm that character now reads in the voice you chose (or the gendered Auto voice).
6. Kokoro first-run download. The first time you use one of Decoupage's own Kokoro voices, it downloads (a progress banner shows while it does). Trigger that — start Read Aloud (or pick a Kokoro voice for a character first, if the picker offers them) and watch for the download banner. Confirm the download completes, the banner goes away, and playback then uses the Kokoro voice. (If the download stalls, errors, or never finishes, note exactly what you saw — this first-run download is going to other machines for the first time.)
7. While playback is running, switch to another tab (e.g. Preview, Cast). Playback should keep going. Return to Write. Try Pause → Resume from each surface.

Note any of these:

• Do the auto-assigned voices sound natural, and are they distinct enough to tell characters apart? (If voices sound like the robotic Samantha-default even with upgraded voices installed, that's worth reporting.)
• Did pinning a voice in the Cast tab actually take effect for that character?
• Did the Kokoro voice download on first use work — did the progress banner appear, complete, and then play in the Kokoro voice? Anything off about it (stalled, errored, confusing)?
• Once Kokoro was downloaded: did its voices sound noticeably better than the Apple system voices?
• If you set a character's gender and left the voice on Auto — did the auto-picked voice (e.g. "Auto — Female voice (Samantha)") actually match the gender when you heard it?
• Did Read From Current Scene actually start at your scene, or did it start elsewhere?
• Did Pause / Resume / Stop behave the way you'd expect?
• With cues no longer spoken, was it still easy to follow who's talking — or did you find yourself wanting the names announced anyway?
• Anything else surprising — wrong pronunciation of a name, weird pacing at scene boundaries, dropouts?

Section 17: Audio Export to .m4a (optional, 15 minutes)

Goal: generate an audio file of the script.

New feature. It uses the same per-character voices as Read Aloud — each character in their own voice, cues not spoken. Generation takes longer than real-time playback because it renders to a file rather than playing live; a long script can take a few minutes.

1. Choose Tools → Export Audio… (or wherever the menu item lives). A save panel should appear.
2. Save to your Desktop as <script-name>.m4a (or whatever Decoupage suggests).
3. Watch the progress indicator while it generates. Cancel mid-export if you want to test that path — confirm cancelling doesn't crash or leave a half-file lying around.
4. Once it finishes, open the resulting .m4a in QuickTime Player or Music.app and play it.
5. Listen for a minute. Confirm: the same per-character voices as Read Aloud (cues not spoken), scene transitions audible (a short pause between scenes), action and dialogue in the right order.

Note any of these:

• How long did the export take for your script's length? (Useful for sizing expectations.)
• Did the cancel path behave cleanly?
• Did the rendered file sound like what you'd heard in Read Aloud, or different?
• Were the pauses between scenes / paragraphs the right length, or too short / too long?
• Anything that came out garbled (mispronunciations, names read as initialisms, etc.)?
• Would you actually use this as-is for a private listen-through of a draft, or does it need more work first?

Section 18: Drafts — version history (important, 10 minutes)

Goal: try Decoupage's automatic version history and the Compare / Restore tools.

Decoupage keeps a version of your document automatically as you work — a history you don't have to manage. New this round, testable by everyone — not tied to any OS version or AI.

1. Make a few rounds of edits to a script, saving as you go, so there are a few versions to look at.
2. Click the clock icon in the toolbar (its tooltip mentions "Drafts timeline — your version history"). The Drafts timeline opens, newest at the top.
3. Try naming or starring a version so the ones that matter stand out.
4. Restore a past version: pick one and restore it. The document text should switch to that version. Notice the footer now shows a confirmation with an Undo — click Undo and confirm your pre-restore text comes back. (The restore is also kept in the timeline, so it stays reversible even after you dismiss.)
5. Compare two drafts: click the Compare toggle in the timeline's toolbar — the rows turn into A/B checkboxes. Pick two drafts (the first is A, the second B), then hit Compare. You should get a scene-grouped diff showing what changed between them, ordered older → newer so additions and removals read naturally. Click Done (or the toggle again) to leave Compare mode.
6. Quick compare against your current draft: on any past version's ⋯ menu, choose "Compare with current draft" — a one-tap diff of that version against what you're working on now (no need to enter Compare mode).

Note any of these:

• Was the timeline easy to find and understand? (Did "automatic version history" land, or did you expect to have to save versions manually?)
• Did Restore + the inline Undo feel safe — i.e. did you trust that you couldn't lose work?
• Did the Compare toggle (A/B picking) feel obvious, or did you have to hunt for it? Was the one-tap "Compare with current draft" useful?
• Was the Compare diff readable? Did the scene grouping help, or get in the way?
• Anything you expected the timeline to do that it didn't?

Section 19: Checkup and Coverage (important, 15 minutes)

Goal: try Decoupage's script health check and — if your Mac supports it — the on-device Coverage read.

What you can test depends on your Mac. The Checkup tab's deterministic pass works on every Mac. Coverage needs macOS 26 with Apple Intelligence enabled. A few AI extras (Coverage's Comparables, Cast's "Read voice from script," Checkup's "Craft checks" group) need macOS 27 — currently a dev beta — so most of you won't see those yet. The steps below mark which is which. Don't treat a missing macOS-27 feature as a bug.

Checkup (every Mac)

1. Switch to the Checkup tab (or View → Show Checkup, ⌘9).
2. Click Run Checkup. You should get page / scene / runtime counts plus a findings list.
3. The findings come back in a labeled group, "Verified checks" — the deterministic counts plus the cast-consistency pass (Checkup reconciles your character cues against the Cast list and flags things like a cue with no casting row, or a name that looks like a variant of another). Click a finding to navigate to where it applies. (On macOS 27 with Apple Intelligence, a second group, "Craft checks," appears below this one — see the macOS-27 steps.)

Coverage (macOS 26 + Apple Intelligence)

4. Open Reports → Coverage (AI)…. If your Mac doesn't have Apple Intelligence, you'll see a card saying Coverage isn't available on this Mac — that's expected; skip to the notes.
5. If it's available, click Generate. After a moment you should get an on-device reader's-pass led by a Premise, then structure, pacing, production scale, strengths, concerns, and per-character notes. The facts card is labeled "From Decoupage's engine — measured, not estimated"; on a long result a sticky bar pins the verdict + facts with a "Jump to" section menu. Everything runs on your Mac — nothing is sent to the cloud.
6. Try Export on the coverage to get a PDF/Markdown version.

macOS 27 only (skip unless you're on the dev beta)

7. (macOS 27) In Coverage, look for a Comparables section — real comparable films / series plus a one-line positioning note. It's empty/absent on macOS 26.
8. (macOS 27) In the Cast tab, open a character and look for "Read voice from script." It produces a transient on-device read of how that character speaks; the result lands in a dashed, tinted box labeled "VOICE READ · AI" with a "transient · not saved" note and a Dismiss button (it isn't saved into the profile). Hidden on macOS 26.
9. (macOS 27) In Checkup, after Run, look for a second findings group, "Craft checks" (badged AI · REVIEW) — on-device per-scene proofreading sitting alongside the deterministic "Verified checks" group. Hidden on macOS 26 (you may see a subtle "available on macOS 27" line instead).

Note any of these:

• Did Checkup's counts match what you expected? Were the cast-consistency findings useful, or noisy?
• If you ran Coverage: did the read feel like a real reader's first pass, or generic? Did it get any facts wrong?
• Was it clear when a feature was unavailable on your Mac (vs. just broken)?
• (macOS 27 testers only) How did Comparables / Read voice from script / Craft checks land?

Section 20: Real Writing (essential, open-ended)

Goal: write something real, or work on something you're already writing.

This is the most important section. Spend 30 minutes or several hours actually writing in Decoupage. Start a new scene or script, or work on something existing.

Note any of these:

• What did you reach for that wasn't there?
• What got in your way?
• What worked smoothly?
• How did Decoupage compare to Final Draft (or whatever you usually use)?
• Did any keyboard shortcuts not match what you expected from other apps?
• Were there any moments where you thought "I wish I could…"?
• Were there any moments where you thought "this is better than what I'm used to"?

On performance: Tab switches, sidebar clicks, every export (PDF, Casting Breakdown, Master Schedule, Audio), and the Preview tab should stay snappy on a 100+ page document. Typing in the Write tab on long scripts may lag a little — a known architectural cost, with a fix scoped for its own focused session. Reporting it is fine, but it won't surprise me. More useful: anything else that feels slow — especially on the production tabs, where snappiness is the bar.

This open-ended section is where the best feedback usually comes from. Don't worry about being systematic. Write, and notice what happens.

When You're Done

Fill out the Findings Log and email it back. No deadline — whenever you're ready.

If you stop midway and decide testing isn't fun anymore, that's useful too. Send back what you have with a note on why you stopped.

Thanks for doing this.