agentleFS
clearsmogclearsmog/claude-skills/doc-review/SKILL.md

doc-review

Skill2 starsChanged 9 months agoMITView on GitHub

Universal file review — auto-detects format and launches appropriate review agents in parallel. Use for quality review of any supported file type (.tex, .qmd, .typ, .py, .md).

15 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Universal Review

Auto-detect file format and launch the right review agents in parallel. Read-only — no fixes applied.

Format Detection

Determine file type from extension:

  • .tex → Beamer LaTeX slides
  • .qmd → Quarto RevealJS slides
  • .typ → Typst document (slides, docs, guides, CVs)
  • .py → Python script
  • .md → Markdown

For .typ files, also detect document type from content:

  • Contains #import "@preview/polylux" or #import "@preview/touying" → Typst slides
  • Contains #import "@local/qk" → Uses qk component library
  • Contains #set page(width: 16cm, height: 9cm) or similar → Presentation
  • Otherwise → Typst document (guide, essay, CV, etc.)

Routing Table

ExtensionAgents (parallel)Condition
.texproofreader + document-auditor + pedagogy-reviewerAlways
.tex+ diagram-reviewerIf TikZ found (\begin{tikzpicture})
.qmdproofreader + document-auditor + writing-style-criticAlways
.qmd+ quality-criticIf .tex sibling exists
.typtypst-reviewer + proofreader + writing-style-criticAlways
.typ+ pedagogy-reviewerIf slide format detected
.pygeneral-purpose subagent with a Python code-review briefAlways
.mdproofreader + writing-style-criticAlways

Steps

  1. Parse $ARGUMENTS — resolve file path
  2. Detect format from extension
  3. For .typ files, read file to detect document subtype
  4. Launch agents in parallel per routing table
  5. Save each agent's returned report to quality_reports/ at the path named in that agent's Output section (read-only agents return their report as text instead of writing it), then collect them
  6. Synthesize a consolidated summary table

Output

# Review Summary: [Filename]

| Dimension | Agent | Critical | Major | Minor | Verdict |
|-----------|-------|----------|-------|-------|---------|
| [varies] | [agent name] | N | N | N | [status] |

## Overall: [EXCELLENT / GOOD / NEEDS WORK / POOR]

## Critical Issues (if any)
[List top issues requiring immediate attention]

## Reports Generated
- quality_reports/[file]_report.md
- quality_reports/[file]_typst_review.md
- ...

Quality Score Rubric

See typst/references/quality-gates.md for the authoritative score rubric, severity definitions, and dispatch table.

Important

  • This is a read-only review — no fixes applied
  • For review + fix, use /finish instead
  • For format-specific deep review, use the individual agents directly

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.