Skip to content
SkillScript-led2,918 tokens in use

short-drama-produce

Runs image, video, voice and music generation for a short drama only after you confirm a previewed batch, with resume support and keys kept outside the project.

4,578 stars on the repository MIT Licensezenstory-ai/drama-skills · skills/short-drama-produce
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-produce costs

207tokens idle
2,918tokens in use
8,191tokens, 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/fixture_adapter.py, scripts/production_tool.py, scripts/provider_adapters.py, scripts/selftest.py.

Touches

shelldetectednetworkdetectedwrites 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 →
Good
  • Frontmatter11/15
  • Structure16/25
  • Examples12/15
  • Output0/15
  • Size13/15
  • Resources8/15

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 →
Good
  • Runs shell commands−5

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

    SKILL.md, line 20
  • Writes files−8

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

    scripts/fixture_adapter.py, line 42
  • Reaches the network−10

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

    scripts/provider_adapters.py, line 27

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity96
  • Maintenance100
  • Community52
  • Maturity20
  • Momentum40

short-drama-produce sends approved prompts to the media generation services you configure and saves the results and run records in the project. It covers images, video clips (single shots and multi-shot groups), voice-over and background music.

Nothing is sent until you confirm. The skill builds a preview of the whole batch: count, target service and model, durations or sizes, and every output path. A confirmation code tied to the exact prompt text, reference files and parameters then unlocks the run, and any edit invalidates it. It does not write prompts, change lines or choose voices. The skill’s text is in Chinese.

What You Get

  • A batch preview listing count, service, model, parameters and output paths
  • Confirmation receipts tied to a code that expires when anything changes
  • Generated images, video clips, voice lines and music saved under the episode’s output folder
  • Run records, a status report and an audit that finds orphaned remote jobs
  • Built-in adapters for GPT Image 2, Seedance, MiniMax H3 video, MiniMax music and MiniMax speech

Common Use Cases

  • Generate every first-frame image for one episode after approving a batch preview
  • Resume a video batch after a dropped connection without paying twice
  • Synthesize the dialogue lines of an episode with fixed voice presets

What Makes It Unique

Its confirmation is part of the tooling. A confirm code is bound to the exact prompt section, reference files, parameters and output paths, a vague ‘go ahead’ does not count, and the code is spent once a remote request is made. Resume and collect steps recover paid jobs after interruptions, and service keys live outside the project folder.

Under The Hood

A bundled Python tool, production_tool.py, prepares batches, records confirmations, runs jobs, checks status, collects results and audits. It reads prompt documents and reference images or audio from the project and writes media files and run records into the episode’s output folder. Once you confirm, it sends requests over the network to the generation service named in an adapter config that lives outside the project, with API keys taken from environment variables or the system credential store, and it polls for results. Those services bill you for each call. It runs only when you ask for generation.

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