agentleFS
CohozenCohozen/magasins-u-mcp-server/.claude/skills/live-check/SKILL.md

live-check

Skill1 starsChanged 20 days agoMITView on GitHub

Run the MCP server locally against the real coursesu.com site and call its tools read-only, to verify a change end to end. Use when asked to test, verify or try a tool "for real", "en vrai", or against the live site.

4 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions4 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 4

    1. NahumLitvin/prismantis★ 426same name
    2. iamrohithrnair/merman★ 42same name
    3. markneonin/paneline★ 10same name
    4. nadavnaveh/skills★ 3same 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 repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 3 are distinctive, and 5 are 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.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Read-only. Never call a write tool (cart, list, slot, order) without the maintainer's explicit approval for that call.

    line 12
  2. Distinctive:

    Never click or solve a Cloudflare challenge.

    line 14
  3. Very specific:

    Keep the default pacing (DELAY_MS), or at least 1500-3000.

    line 16
  4. Very specific:

    Run it in the background and keep its PID.

    line 30
  5. Very specific:

    If the log says human action required with needs_login, ask the maintainer to sign in in the Chromium window that opened (on macOS there is …

    line 33
  6. Distinctive:

    Call tools with the bundled client:

    line 37
  7. Very specific:

    Check the answers against what the tool promises (fields present, no nulls where data is expected), and the log for task failed lines.

    line 46
  8. Very specific:

    Stop the server with kill -TERM <pid> on the Node PID itself (not pkill -f, which also hits the wrapping shell).

    line 49

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Live check

Verifies the server against the real site, read-only, on the developer's machine.

Rules

  • Read-only. Never call a write tool (cart, list, slot, order) without the maintainer's explicit approval for that call.
  • Never click or solve a Cloudflare challenge. On a challenge, a login wall or an HTTP 403/429, stop and report.
  • Keep the default pacing (DELAY_MS), or at least 1500-3000. A live check is a handful of calls, not a loop.
  • Tool output contains personal data (orders, receipts, lists): read it, summarise it, never write it into the repository. Scratch files go to a scratch directory or .local/.

Steps

  1. Build-free start, with a throwaway token and the local data directory (git-ignored):

    MCP_TOKEN=local-check-0123456789abcdef HOST=127.0.0.1 PORT=8765 DELAY_MS=1500-3000 \
      node src/server.ts > "$SCRATCH/server.log" 2>&1 &

    Run it in the background and keep its PID. Wait for the startup probe or task line with "label":"probe" in the log.

  2. If the log says human action required with needs_login, ask the maintainer to sign in in the Chromium window that opened (on macOS there is no noVNC). The server logs session restored on its own once signed in; do not poll the site yourself.
  3. Call tools with the bundled client:

    export MCP_URL=http://127.0.0.1:8765/mcp MCP_TOKEN=local-check-0123456789abcdef
    node scripts/call.ts                       # list tools
    node scripts/call.ts session_status
    node scripts/call.ts search_products '{"query":"lessive liquide"}'
  4. Check the answers against what the tool promises (fields present, no nulls where data is expected), and the log for task failed lines.
  5. Stop the server with kill -TERM <pid> on the Node PID itself (not pkill -f, which also hits the wrapping shell). Expect exit code 0 and a fresh data/session.json.

After the check

  • data/ holds a live session of the maintainer's account: mention it, and offer to delete it when the session is no longer needed.
  • Report what was verified live and what was not.

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.