agentleFS
nigo-studionigo-studio/ai-film-skills/CLAUDE.md

ai-film-skills

CLAUDE.md15 starsChanged 3 months agoMITView on GitHub

Two production skills from the pipeline behind CROW LINE, an AI-generated action film. This folder is a ready-to-use workspace: the skills live in .claude/skills/ and trigger automatically once this folder is opened in Claude Code. Always read the matching SKILL.md before writing, and follow its steps and self-checks exactly. The two skills compose: build the character sheet first with character-consistency, then hand that sheet to video-prompt-director as one of the reference images. When the user comes back with a generated…

The file

We split this CLAUDE.md into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is 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.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Always read the matching SKILL.md before writing, and follow its steps and self-checks exactly.

    line 12
  2. Distinctive:

    When the user comes back with a generated take and points at what looks wrong or different from what they pictured:

    line 22

≈ “Common” is an estimate: at least 1 in 5 of the 28,029 CLAUDE.md 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

AI Film Skills — Starter Kit

Two production skills from the pipeline behind CROW LINE, an AI-generated action film. This folder is a ready-to-use workspace: the skills live in .claude/skills/ and trigger automatically once this folder is opened in Claude Code.

The two skills

SkillOpen it when
video-prompt-directorThe user gives a scene direction, action choreography, or shot idea and wants a full Seedance 2.5 reference-to-video prompt built from it.
character-consistencyThe user wants one character to stay identical across generated shots — expanding a character brief, building the ultra-detailed close-up, or assembling the three-panel character sheet.

Always read the matching SKILL.md before writing, and follow its steps and self-checks exactly. The two skills compose: build the character sheet first with character-consistency, then hand that sheet to video-prompt-director as one of the reference images.

Ground rules

  • Generation prompts are always written in English, whatever language the user speaks. Talk to the user in their language.
  • These skills write prompts and nothing else. The user runs the generation on their own platform and picks the takes. Never claim to have generated media, and never call a generation API on your own initiative.
  • The direction belongs to the user: who appears, where, what happens, which moment matters most. Never override it. The prompt craft belongs to the skill.

Iterating on a take

When the user comes back with a generated take and points at what looks wrong or different from what they pictured:

  1. Find the prompt sentence that caused it — or the sentence that is missing.
  2. Fix that part only. Keep the rest of the prompt and the template structure intact.
  3. To remove something that appeared in frame, delete the sentence that put it there. Adding "no X" injects the token X and tends to backfire.
  4. Deliver the complete revised prompt, ready to paste — not a diff, not a commentary.

Layout

  • .claude/skills/video-prompt-director/ — scene direction in, submit-ready Seedance 2.5 prompt out
  • .claude/skills/character-consistency/ — character brief in, three-panel identity sheet out
  • README.md — install and usage notes for humans

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.