screenplay-review-checklist

zhlmi/screenplay-review-checklist

screenplay-review-checklist is an agent skill project by zhlmi on GitHub: Adversarial script pitfall diagnosis: uses 37 criteria to look only for problems in your story. It covers review. 13 stars, last updated 2026-08-15, no license declared.

Editor's notes

The 37 criteria fall into seven layers, A to G, with findings graded at three severity levels, and form-adaptation rules R-01 to R-07 keep legitimate techniques in exported long role-play sessions from being flagged as defects. On start it must ask which review mode to use: if an AI has already rated or helped write the story, it recommends a first pass by an independent model as a sub-agent, after which the main model checks each piece of evidence against line numbers. The report is a Markdown file and ends by inviting the author to push back. Some criteria are drawn from He Yi's 168 screenwriting principles (《影视剧写作心法 168 条》), and the repo includes the full original text with the author's permission.

Type
Screenwriting: scripts · adaptation · review
Stages
Review
Format
Agent skill
Language
Chinese
GitHub stars
★ 13
Last updated
2026-08-15
License
None declared
SKILL.md files
1
Checked by hand
2026-10-13

In the author's words

An "adversarial screenplay/story landmine diagnosis" skill for AI agents. It has the model work through a 37-criterion checklist and only find faults in your story, with no praise and evidence for every item, then invites the author to push back at the end, turning landmines into load-bearing walls.

Translated from the Chinese original.

Skills in this repo (1)

Names and descriptions come from each SKILL.md frontmatter; Chinese descriptions are shown in our English translation.

Install

npx skills add zhlmi/screenplay-review-checklist

# pick agents (claude-code / codex / cursor / opencode / gemini-cli / openclaw)
npx skills add zhlmi/screenplay-review-checklist -a claude-code -a codex

Default commands use the skills CLI and were not verified one by one; follow the repo README. Read SKILL.md before installing. Install guide →

Similar projects