Skip to content
SkillScript-led1,729 tokens in use

short-drama-breakdown

Measures benchmark short dramas for episode and shot length, dialogue density, openings and endings, then compares them with your own plan or cut.

4,570 stars on the repository MIT Licensezenstory-ai/drama-skills · skills/short-drama-breakdown
npx skills add zenstory-ai/drama-skills -y -g
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing short-drama-breakdown costs

356tokens idle
1,729tokens in use
15,202tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Script-led

Script-ledMixedModel-led

Most of this skill is scripts, so it behaves much the same whichever model runs it.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: scripts/breakdown_tool.py, scripts/selftest.py.

Touches

shelldetectedwrites filesdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter11/15
  • Structure8/25
  • Examples9/15
  • Output0/15
  • Size11/15
  • Resources11/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 14
  • Writes files−8

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 34

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity100
  • Maintenance100
  • Community52
  • Maturity20
  • Momentum40

short-drama-breakdown studies reference series or your own finished cuts. A script measures episode length, shot length and dialogue density from the videos, and the agent reads sampled frames and subtitles to describe the opening beat, how information is delivered, the end-of-episode hook and how one episode leads into the next.

It lays your plan or your own cut next to the benchmark in a gap table, hands the gaps to the skill that owns each fix, and can reverse-engineer a video into a verbatim script and, if asked, setup documents, storyboard and prompts. Reverse-engineered material stays in the benchmark folder unless the work is yours or licensed. The skill’s text is in Chinese.

What You Get

  • A registry file per reference series with its source and permitted use
  • Measured statistics: episode length, shot length, dialogue density, opening and ending timing
  • A written reading of each reference covering opening, information delivery, hooks and cross-episode links
  • A same-ruler gap table between the benchmark and your own plan or cut
  • Optionally a reconstructed script and setup, storyboard and prompt documents from a video you own

Common Use Cases

  • Study what top series do in their first three seconds before writing your own opening
  • Measure your finished cut against a benchmark and send each gap to the right stage skill
  • Reverse a video you own into a script to remake or continue it

What Makes It Unique

It keeps measurement, interpretation and unknowns apart in every note, and it keeps study material fenced off: real titles, character names and lines from a study reference may not enter your own project documents. Reconstructed documents leave the benchmark folder only when the work is registered as owned or licensed with the basis written down. It never sets pacing numbers for you; the creator confirms them.

Under The Hood

The agent runs a bundled Python script on the local machine with ffmpeg and ffprobe to measure videos, cut contact sheets of representative frames, attach subtitle transcripts and optionally cut shot slices, writing the results under the project’s input folder. It reads the videos and subtitles you provide and writes Markdown and JSON files in the project. The instructions ask whether footage may go to an external model service and keep records of that choice. It does not fetch web pages or call web services by itself.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers