book-screenplay
Skill40 starsChanged 2 months agoApache-2.0View on GitHub
Screenplay craft reference: format, three-act and sequence structure, scene construction, dialogue subtext, A/B/C story weaving, series bibles, and screenplay-specific AI tells. Read by all agents on film, TV, and web-series projects.
In this repository23 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- dashboard/CLAUDE.mdCLAUDE.md
- .cursor/rules/velith-editing.mdcCursor rule
- .cursor/rules/velith-genres.mdcCursor rule
- .cursor/rules/velith-pipeline.mdcCursor rule
- book-academicSkill
- book-draftSkill
- book-editSkill
- book-fictionSkill
- book-gameSkill
- book-genre-creatorSkill
- book-ideationSkill
- book-illustrateSkill
- book-initSkill
- book-nonfictionSkill
- book-outlineSkill
- book-poetrySkill
- book-publishSkill
- book-statusSkill
- book-technicalSkill
- book-visualsSkill
- loomSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 60
Every scene starting with the character entering and ending with them leaving.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Screenplay
A screenplay is a document for collaborators, read by people who read hundreds. It must be fast, visual, and impossible to put down on page 10. Novelistic interiority does not exist here; everything the audience learns, they see or hear.
Structure
- Three acts by page: feature 90-120 pages; Act 1 to ~25, Act 2 to ~85, Act 3 to end. One page ≈ one minute.
- Eight sequences (two per act in 1 and 3, four in 2), each 12-15 pages with its own mini-arc: status quo → disruption → strategy → complication → crisis → decision → new status quo. Sequences prevent the second-act sag.
- Beat placement: inciting incident by page 10-12; first-act turn ~25; midpoint ~55 (a reversal, not an event); low point ~75; third-act turn ~85.
- TV: cold open, act breaks at commercial points (4-5 acts for network, 3 for streaming), A/B/C stories per episode, season arc bible with per-episode outlines. Pilot must establish the engine (what generates episodes forever).
- Web series: 8-15 pages per episode, hook in the first 30 seconds, cliffhanger every episode.
- Short: 5-30 pages, one turn, no subplot.
Scene
- Slug line:
INT./EXT. LOCATION - DAY/NIGHT. Location names consistent across the script (bible). - Every scene: someone wants something, something is in the way, the scene ends with the situation changed. Outcome: yes-but, no-and, no-but.
- Enter late, leave early. Cut the arrival, the greeting, the goodbye.
- Action lines: present tense, only what can be seen or heard, 3-4 lines max per paragraph, white space is pacing. No "we see." No camera directions unless writer-director.
- Character introduction: NAME in caps at first appearance, age, one telling detail. Not a paragraph of backstory.
Dialogue
- Subtext over statement. Characters talk around the thing.
- Distinct voice per character: rhythm, vocabulary, what they never say. Cover the names; the reader should still know who is speaking.
- Monologues ≤ 5 lines unless the script earns a set piece.
- Parentheticals rarely; the actor decides.
- V.O. and O.S. sparingly and consistently.
- Korean: 존댓말/반말 relationships fixed per pair in the bible; a shift is a story event. 지문 in 현재형.
A/B/C stories
A-story (main plot, ~70%), B-story (relationship or theme, ~25%), C-story (runner, ~5%). B carries the theme; A carries the plot; they collide at the midpoint and the climax. Alternate A and B every 2-3 scenes; do not run five A scenes in a row.
Format
Industry-standard layout (Courier 12, 1.5" left margin, dialogue block width). Drafts are markdown with a strict convention so book-publish can export to Fountain and PDF:
INT. KITCHEN - NIGHT
Action line.
CHARACTER
(parenthetical)
Dialogue.Transitions only between sequences. No bold, no italics except in the title page.
Screenplay-specific tells
- Action lines that describe emotion ("she feels devastated").
- Characters stating their motivation or the theme.
- Every scene starting with the character entering and ending with them leaving.
- Uniform scene length.
- Dialogue that answers the previous line directly.
- On-the-nose exposition disguised as argument ("As you know, we've been married ten years").
- Montages to skip the hard scene.
- The final scene that mirrors the first scene with one word changed.
Bible
Per character: verbal signature, want/need, 존댓말 map. Per location: what it looks like once, then consistent. Props and costume that carry meaning are tracked. Time of day and day count across the script.
Validation (architect scoring for screenplay)
Beat placement within tolerance; each sequence has a mini-arc; A/B collide at midpoint and climax; every scene has a change; page count in range for the format; no scene without a slug; character introductions correct; dialogue distinguishable by voice.
Visuals
Scripts carry no interior images. Pitch and series-bible packages do: cover/key art, character reference sheets, location boards, a tone board of 6-9 images, and a season-arc diagram (figure-engineer). All through /book-visuals; the art bible for a pitch deck is the show's visual identity.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

