qa
Skill2 starsChanged 9 months agoMITView on GitHub
Adversarial quality audit loop. Critic finds issues, fixer applies fixes, loops until APPROVED (max 5 rounds). Works with any document format.
In this repository23 other files it gives its agents.
- compileSkill
- context-statusSkill
- create-documentSkill
- deploySkill
- devils-advocateSkill
- diagramSkill
- doc-reviewSkill
- excellenceSkill
- extract-diagramsSkill
- finishSkill
- google-tasksSkill
- image-searchSkill
- learnSkill
- mindmapSkill
- pedagogy-reviewSkill
- proofreadSkill
- sec-edgarSkill
- translateSkill
- typstSkill
- validate-bibSkill
- verifySkill
- visual-auditSkill
- waybackSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 39
Check freshness (re-render if QMD newer than HTML)
- Distinctive: line 40
Verify TikZ SVGs if applicable
- Distinctive: line 48
Check for auxiliary files (.aux, .bbl)
- Very specific: line 56
If not APPROVED, launch quality-fixer agent to apply fixes (Critical → Major → Minor), re-compile, and verify.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Adversarial Quality Audit Workflow
Iterative critic/fixer loop for any supported document format.
Format Detection
Determine file type from extension:
.tex→ Beamer LaTeX slides.qmd→ Quarto RevealJS slides (compare vs.texbenchmark if it exists).typ→ Typst document (audit against quality standards)
Workflow
Phase 0: Pre-flight → Phase 1: Critic audit → Phase 2: Fixer → Phase 3: Re-audit → Loop until APPROVED (max 5 rounds)Hard Gates (Non-Negotiable, Universal)
| Gate | .tex | .qmd | .typ |
| Overflow | No overfull hbox > 10pt | No content cut off | No content exceeds page |
| Content Quality | No undefined citations | Interactive >= static plots | All imports resolve |
| Typography | Consistent formatting | Notation fidelity | Font consistency |
Phase 0: Pre-flight
For .qmd files:
- Locate Beamer (.tex/.pdf) and Quarto (.qmd/.html) files
- Check freshness (re-render if QMD newer than HTML)
- Verify TikZ SVGs if applicable
For .typ files:
- Locate the file, check it compiles
- Detect document type (slides, CV, guide, essay, report)
For .tex files:
- Locate the file, verify it compiles
- Check for auxiliary files (.aux, .bbl)
Phase 1: Initial Audit
Launch the quality-critic agent to audit the document. Save the report it returns to quality_reports/[FILENAME]_qa_critic_round1.md (re-audits: _round[N].md) before launching the fixer.
Phase 2: Fix Cycle
If not APPROVED, launch quality-fixer agent to apply fixes (Critical → Major → Minor), re-compile, and verify.
Phase 3: Re-Audit
Re-launch critic to verify fixes. Loop back to Phase 2 if needed.
Iteration Limits
Max 5 fix rounds. After that, escalate to user with remaining issues.
Final Report
Save to quality_reports/[FILENAME]_qa_final.md with:
- Hard gate status
- Iteration summary (rounds, issues fixed per round)
- Remaining issues (if any)
- Final verdict
Quality Gates
See typst/references/quality-gates.md for score rubric, commit thresholds, and severity definitions.
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
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.

