agentleFS
caiopizzolcaiopizzol/tres-pinos/.agents/skills/write-guide/SKILL.md

write-guide

Skill7 starsChanged 33 days agoMITView on GitHub

Writes or updates a connector's setup guide in the Três Pinos repository, following docs/GUIDE.md, with its GIFs. Use when asked to document how to set up a connector, or to add a new connector's README.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 0

    Distinctive

    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
  1. Very specific:

    Read docs/GUIDE.md and the connector's code: src/server.mjs for the settings, src/tools.mjs for the tools and src/guard.mjs for what it …

    line 10
  2. Very specific:

    Write each section from what the code does.

    line 12
  3. Very specific:

    If the connector has a sign-in page (its server.mjs passes PUBLIC_URL to serve()), record it:

    line 14
  4. Very specific:

    Look at every frame of every GIF before using it.

    line 37

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “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

Write a connector guide

The guide is connectors/<connector>/README.md, in Portuguese, with the sections of docs/GUIDE.md, in order.

  1. Read docs/GUIDE.md and the connector's code: src/server.mjs for the settings, src/tools.mjs for the tools and src/guard.mjs for what it refuses.
  2. Write each section from what the code does. Never guess a menu name, a setting or a behavior; if you can't find it, write <!-- TODO: ... --> and say so.
  3. If the connector has a sign-in page (its server.mjs passes PUBLIC_URL to serve()), record it:

    node .agents/skills/write-guide/scripts/record-sign-in.mjs <connector> connectors/<connector>/entrar.gif

    It runs the real connector with fictional data. It needs ffmpeg, and Chromium for Playwright, installed once with bunx playwright install chromium.

  4. For the service's own screens, such as a bank's, ask the owner to log in, then go through the steps in their browser and take a screenshot of each screen. Write the steps from the screenshots, and join them into a GIF:

    sh .agents/skills/write-guide/scripts/screens-to-gif.sh connectors/<connector>/<step>.gif <screen.png>...

    Park the mouse outside the area you capture before each screenshot, or it ends up in the image. The owner types every password and code. Stop and ask before anything that creates, accepts or downloads, and never capture or repeat a key, secret, balance, CNPJ, account number or name.

    To animate it, with a cursor clicking each step, see connectors/inter/guide/: a HyperFrames project over the same screenshots. Copy it and change the clicks in index.html.

  5. Look at every frame of every GIF before using it. Run bun run check and the connector's tests.

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.