docs-change
Skill70 starsChanged 7 days agoApache-2.0View on GitHub
Prepare a docs-only cap-evolve PR for review - no real run, no screenshots, just a local link check before going straight to review-cycles. Use when every changed file is a .md doc, site/**/*.html prose, a SKILL.md body, or comments-only. If any code, adapter, or dashboard-frontend file changed too, use testing or dashboard-change for those files instead.
In this repository27 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- llms.txtllms.txt
- dashboard-changeSkill
- open-prSkill
- review-cyclesSkill
- testingSkill
- agent-optimizeSkill
- evographSkill
- gepaSkill
- hill-climbSkill
- skilloptSkill
- mcp-toolSkill
- skill-packageSkill
- system-promptSkill
- toolsSkill
- wikiSkill
- run-optimizerSkill
- orchestrateSkill
- using-cap-evolveSkill
- baselineSkill
- diagnoseSkill
- evaluateSkill
- finalizeSkill
- gateSkill
- implement-and-checkSkill
- intakeSkill
- reportSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Check the diff's file list: *.md, docs/, site/ prose, .github/.md, or a SKILL.md/CLAUDE.md body — and nothing under core/, …
- Very specific: line 30
Check ci.yml's docs-links job for the current args if this skill's copy drifts.
- Very specific: line 35
If the change touched CONTRIBUTING.md, README.md, or any SKILL.md's documented command, also spot-check that the command still exists: grep …
- Very specific: line 41
Don't dispatch a benchmark workflow and don't take dashboard screenshots for a docs-only change — there's no behavior to demonstrate.
≈ “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
Docs change
Pure documentation changes skip both real runs and screenshots — the only thing to verify is that nothing broke.
1. Confirm the change really is docs-only
Check the diff's file list: *.md, docs/**, site/** prose, .github/**.md,
or a SKILL.md/CLAUDE.md body — and nothing under core/, skills/*/*/scripts,
dashboard/, ci/benchmarks/lib/, or any benchmark adapter code. If anything else
changed, route that part through testing or dashboard-change and only use this
skill for the doc-only portion.
2. Link check (the only check that matters here)
Same command CI's docs-links job runs:
lychee --offline --no-progress --include-fragments \
--exclude-path benchmarks/tau2/examples/airline/run_full \
--exclude-path benchmarks/skillsbench/examples/shared_skills/run_full \
"README.md" "docs/**/*.md" "benchmarks/**/*.md" "*.md"Check ci.yml's docs-links job for the current args if this skill's copy drifts.
Zero errors required. A relative link that moved (renamed file, moved directory) is the most common failure — fix the link, don't exclude the path.
If the change touched CONTRIBUTING.md, README.md, or any SKILL.md's
documented command, also spot-check that the command still exists:
grep it against core/cap_evolve/cli.py's subcommand list, or just run it.
3. No real run, no screenshots
Don't dispatch a benchmark workflow and don't take dashboard screenshots for a
docs-only change — there's no behavior to demonstrate. Go straight to
review-cycles.
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.

