monarch-initiative/mechmaker/review-record
- community
Checking that the SKILL.md file is still in the repository.
Audit one IngestMech record, or a pull request that changes records, without editing anything. Use when asked to review, check, audit or assess an ingest record or a curation pull request. Produces findings; changes nothing.
Install
npx vskill@latest install monarch-initiative/mechmaker/review-recordSKILL.md
Review a record
Read only. The output is a list of findings, each with a location and a reason. Fixes belong to curate-record, in a separate step, when asked.
Checks, in order
- Machine checks.
just validate data/ingests/<stem>.yaml. Every
error is a finding. A clean run proves the citations exist, the quotes are exact and the terms are real. It proves nothing else.
- Scope. Is this one ingest by the rules in
docs/DOMAIN.md?
Is it a duplicate of another record?
- Quote fit. For each evidence item, read the snippet against the claim
it is attached to. A real quote under the wrong claim passes validation and is still wrong. This is the check machines cannot do.
- Term fit. Is each bound term the most specific accurate one? Is a
broad binding explained in notes?
- Direction. Does
supportsmatch what the quote says? - Description. Take each fact the
descriptionstates (a date, a
place, a use, a number). Each needs a quote: in description_evidence, or in the section that states the same fact. A fact with neither is an error, even when it is true.
- Gaps. What does the record claim with no evidence? What is missing
that the sources would support?
Output
A table: location (section and item), finding, severity (error, warning, note), and the evidence for the finding. End with a verdict: approve, approve with notes, or changes needed. If you were asked to record the review, add a history record with --event REVIEW, and --outcome no_change when you changed nothing.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingexample/ingestmech/.claude/skills/review-record/SKILL.md
Popularity
- Repository stars
- 2
- 7-day trend score
- 10
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/monarch-initiative/mechmaker/review-record)