agentleFS
raniellimontagnaraniellimontagna/agent-toolkit/AGENTS.md

agent-toolkit

AGENTS.md3 starsChanged 32 days agoMITView on GitHub

This project has a knowledge graph at graphify-out/ with god nodes, community structure, and cross-file relationships. When the user types /graphify, invoke the skill tool with skill: "graphify" before doing anything else.

The file

Show its rules and text

We split this AGENTS.md into the rules it gives an agent

14 rules. About 0 are common, 6 are distinctive, and 8 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 8

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Prefer rtk as the command prefix when it is available, for example rtk pnpm run check.

    line 7
  2. Very specific:

    If rtk is not installed yet, run the equivalent command without the prefix or install RTK through this toolkit.

    line 8
  3. Distinctive:

    Keep these project rules portable.

    line 9
  4. Distinctive:

    Use Conventional Commits for every commit message.

    line 13
  5. Distinctive:

    Keep commit subjects concise and scoped, for example:

    line 14
  6. Very specific:

    Do not create commits with vague subjects such as update, changes, fixes, or wip.

    line 19
  7. Distinctive:

    Before claiming work is complete, run lint and tests.

    line 23
  8. Distinctive:

    Prefer the full check whenever the change touches code, scripts, package metadata, installer behavior, tests, or project rules:

    line 24
  9. Distinctive:

    If any verification command fails, fix the issue or report the exact blocker before saying the task is done.

    line 37
  10. Very specific:

    When the user types /graphify, invoke the skill tool with skill: "graphify" before doing anything else.

    line 43
  11. Very specific:

    If graphify-out/wiki/index.md exists, use it for broad navigation instead of raw source browsing.

    line 49
  12. Very specific:

    Read graphify-out/GRAPH_REPORT.md only for broad architecture review or when query/path/explain do not surface enough context.

    line 51
  13. Very specific:

    If queries or the HTML graph surface stale project names, deleted paths, or legacy files, treat graphify-out/ as stale cache.

    line 53
  14. Very specific:

    After modifying code, run graphify update .

    line 54

≈ “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

--- project-doc ---

Agent Toolkit Project Rules

Shell Commands

  • Prefer rtk as the command prefix when it is available, for example rtk pnpm run check.
  • If rtk is not installed yet, run the equivalent command without the prefix or install RTK through this toolkit.
  • Keep these project rules portable; do not reference machine-local instruction files.

Git Workflow

  • Use Conventional Commits for every commit message.
  • Keep commit subjects concise and scoped, for example:
    • feat: add graphify installer
    • fix: resolve compiled package root
    • docs: update project rules
    • test: cover skill scope filtering
  • Do not create commits with vague subjects such as update, changes, fixes, or wip.

Verification

  • Before claiming work is complete, run lint and tests.
  • Prefer the full check whenever the change touches code, scripts, package metadata, installer behavior, tests, or project rules:
rtk pnpm run check
  • For very small documentation-only changes, at minimum run:
rtk pnpm run lint
rtk pnpm test
  • If any verification command fails, fix the issue or report the exact blocker before saying the task is done.

Graphify

This project has a knowledge graph at graphify-out/ with god nodes, community structure, and cross-file relationships.

When the user types /graphify, invoke the skill tool with skill: "graphify" before doing anything else.

Rules:

  • For codebase questions, first run graphify query "<question>" when graphify-out/graph.json exists. Use graphify path "<A>" "<B>" for relationships and graphify explain "<concept>" for focused concepts. These return a scoped subgraph, usually much smaller than GRAPH_REPORT.md or raw grep output.
  • Dirty graphify-out/ files are expected after hooks or incremental updates; dirty graph files are not a reason to skip Graphify. Only skip Graphify if the task is about stale or incorrect graph output, or the user explicitly says not to use it.
  • If graphify-out/wiki/index.md exists, use it for broad navigation instead of raw source browsing.
  • For visual navigation in this repo, prefer graphify-out/GRAPH_TREE.html over graphify-out/graph.html. The force-directed graph can be too dense because bundled skills dominate the corpus. Generate or refresh the tree with rtk graphify tree --label agent-toolkit.
  • Read graphify-out/GRAPH_REPORT.md only for broad architecture review or when query/path/explain do not surface enough context.
  • Graphify is pinned through tools.lock.json. If the CLI version, installed Graphify skill version, or command warnings do not match the pinned version, update the local Graphify install and reinstall the Graphify skill for the active runtime before trusting fresh graph output.
  • If queries or the HTML graph surface stale project names, deleted paths, or legacy files, treat graphify-out/ as stale cache. Move the old graphify-out/ aside, run rtk graphify update . --force, then regenerate GRAPH_TREE.html.
  • After modifying code, run graphify update . to keep the graph current.

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.