agentleFS
guardanaguardana/guardana/.claude/skills/site-check/SKILL.md

site-check

Skill225 starsChanged 40 days agoApache-2.0View on GitHub

Check the landing page and the generated documentation site (site/, guardana.dev) in a real browser — confirm a page renders after a docs change, review it at phone width, read the console and network for anything that contradicts the no-telemetry promise, collect screenshots as evidence. Use after any change under site/, docs/ or scripts/sitegen/, and before a push that deploys the site.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. Moh4696/open-3d-skills★ 391same name
    2. smilemino/unchore-defend★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository16 other files it gives its agents.

The file

Files
SKILL.md

Site check in a real browser

Target: $ARGUMENTS

Delegate the clicking to the browser agent (page snapshots are large); keep its findings here. Tools: the Playwright MCP (browser_navigate, browser_snapshot, browser_resize, browser_console_messages, browser_network_requests, browser_take_screenshot).

Serve it locally — what Cloudflare serves, byte for byte

uv run python scripts/build_site.py            # rebuild site/docs/ from docs/ (or --check)
python3 -m http.server -d site 8099            # http://localhost:8099/
pagewhat to check there
/the landing page: the counts (sync_site.py --check says they are current), the terminal demo, every link resolves
/docs/the documentation home: the sidebar mirrors docs/index.md, every section has its pages
/docs/usage-scan.html (any usage page)fenced commands render as code, tables fit at 400 px, the status badge matches the page's front matter
/docs/generated/rule-catalog.html and one rule page from the explorerthe rule count, taxonomy links, no "do not edit" marker leaking into the page

Live (https://guardana.dev): look only. There is nothing to log into and nothing to submit.

What a pass covers

  1. The page renders with its real content — no empty section, no raw markdown, no missing generated block. A blank where a count belongs is a finding, not an empty state.
  2. Console errors and failed requests after each page. Any request to a host other than the page's own is a finding: the site promises no telemetry and ships a closed Content-Security-Policy in site/_headers.
  3. Desktop and 400 px: nothing cut off, no horizontal page scroll, code blocks scroll inside themselves, the sidebar collapses.
  4. Keyboard: links reachable in order, visible focus.
  5. Screenshots of the changed pages (before/after when fixing a bug) saved to the scratchpad; attach the paths to the report, and to the commit body when they are evidence.

A site change is not done until uv run python scripts/build_site.py --check and uv run python scripts/generate_llms_txt.py --check are clean — the hook refuses a push to main otherwise, because the push is the deploy.

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.