agentleFS
DataSLDataSL/datasl/AGENTS.md

datasl

AGENTS.md0 starsChanged 18 days agoMITView on GitHub

Instructions for AI coding agents (Claude Code, Codex, Cursor, Copilot, Gemini CLI and others) working in this repository or writing code that uses the Data SL Consulting API. The official, public toolkit of Data SL Consulting, a Paris data and AI consultancy: Run the offline suites before any change. Keep live tests to a handful of requests (the site is a small WordPress install; never send bursts of concurrent requests). Version numbers live in js/package.json, python/pyproject.toml and python/src/datasl/__init__.py; bump all…

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository3 other files it gives its agents.

The file

We split this AGENTS.md 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
  1. Very specific:

    Keep both packages dependency-free and behaviourally identical: same commands (offres, offre, cabinet, contact, chercher, lire, mcp), same …

    line 25
  2. Distinctive:

    Never hard-code prices or offer lists: they come from the API, which is the source of truth.

    line 27
  3. Distinctive:

    Do not add telemetry, analytics or calls to any host other than data-slconsulting.com.

    line 30
  4. Distinctive:

    Run the offline suites before any change.

    line 40

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
AGENTS.md

AGENTS.md

Instructions for AI coding agents (Claude Code, Codex, Cursor, Copilot, Gemini CLI and others) working in this repository or writing code that uses the Data SL Consulting API.

What this repository is

The official, public toolkit of Data SL Consulting, a Paris data and AI consultancy:

  • js/: the npm package datasl (CLI + JavaScript client, Node 18+, no dependencies).
  • python/: the PyPI package datasl (CLI + Python client, Python 3.9+, no dependencies).
  • skills/datasl/SKILL.md: an Agent Skill that teaches an agent to look up the consultancy's offers, prices and content.
  • plugin.json + mcp.json (Agent Plugins format) and .claude-plugin/ + .mcp.json (Claude Code): plugin manifests that bundle the skill with the public MCP server.

The API these packages wrap

  • Base URL: https://data-slconsulting.com/wp-json/datasl/v1. Read-only, public, no authentication.
  • Contract: OpenAPI 3.1 at https://data-slconsulting.com/openapi.json. Documentation: https://data-slconsulting.com/developers/.
  • Routes: GET /offres (filters cible, categorie, champs), GET /offres/{slug}, GET /cabinet, GET /contact; search via GET /wp-json/wp/v2/search?search=....
  • MCP server: https://data-slconsulting.com/api/mcp-server (Streamable HTTP; tools get_offers, search_site, fetch_page).
  • Errors: JSON {"code", "message", "data": {"status", "hint"}}. Surface hint to the user; it says how to fix the call.
  • Rate limit: 120 requests per minute per IP, announced by RateLimit-Policy and RateLimit headers; on 429, honour Retry-After.

Rules when changing code here

  • Keep both packages dependency-free and behaviourally identical: same commands (offres, offre, cabinet, contact, chercher, lire, mcp), same --json output, same exit codes (0 success, 1 API or network error, 2 usage error).
  • Validate inputs before any network call (offer slugs match ^[a-z0-9-]+$; lire only reads pages of data-slconsulting.com, never wp-admin, wp-json, wp-login or /api/).
  • Never hard-code prices or offer lists: they come from the API, which is the source of truth.
  • Decode HTML entities in search result titles (the WordPress search API returns them encoded).
  • User-facing text is in French; code identifiers follow the existing French naming.
  • Do not add telemetry, analytics or calls to any host other than data-slconsulting.com.

Tests

cd js && npm test                       # offline, against a local mock server
cd python && PYTHONPATH=src python -m unittest discover -s tests   # offline
cd python && DATASL_LIVE=1 PYTHONPATH=src python -m unittest tests.test_live   # a few requests to the real site

Run the offline suites before any change. Keep live tests to a handful of requests (the site is a small WordPress install; never send bursts of concurrent requests).

Releasing

Version numbers live in js/package.json, python/pyproject.toml and python/src/datasl/__init__.py; bump all three together. Packages are published by Data SL Consulting from its own npm and PyPI accounts.

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.