agentleFS
ECG-Rudy-GltECG-Rudy-Glt/RELAI/AGENTS.md

RELAI

AGENTS.md2 starsChanged 10 days agoApache-2.0View on GitHub

Start here: 0. Fresh clone? The tasks live in Beads, outside git: run bd bootstrap --yes once to load them from .beads/issues.jsonl (needs bd). Without it bd list is empty. 1. Run relai handoff (current state, next tasks, decisions, pitfalls — a few hundred tokens). relai not installed or not found? Read .relai/notes.md and relai.yaml directly instead — they are plain files, no CLI required. 2. Before building something another project might already provide, check relai find <capability> (or read…

The file

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

8 rules. About 0 are common, 2 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

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

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

    Run relai handoff (current state, next tasks, decisions, pitfalls — a few hundred tokens).

    line 6
  2. Very specific:

    Before building something another project might already provide, check relai find <capability> (or read a shared .relai/index.json …

    line 9
  3. Distinctive:

    Keep this file under 100 lines.

    line 16
  4. Very specific:

    After creating, updating or closing tasks, run scripts/sync-beads.sh and commit .beads/issues.jsonl: that file is the copy of the task list …

    line 20
  5. Very specific:

    Every bd run rewrites ~3 MiB of its database on disk: never run it on a timer or in a loop.

    line 22
  6. Very specific:

    Before committing: gofmt -l .

    line 26
  7. Very specific:

    Commit messages: conventional style (feat:, fix:, docs:…), no Co-Authored-By or "Generated with" trailers.

    line 28
  8. Distinctive:

    Keep it easy: every feature must work for one person with only git, no server.

    line 32

≈ “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

relai — instructions for AI agents

Start here: 0. Fresh clone? The tasks live in Beads, outside git: run bd bootstrap --yes once to load them from .beads/issues.jsonl (needs bd). Without it bd list is empty.

  1. Run relai handoff (current state, next tasks, decisions, pitfalls — a few hundred tokens). relai not installed or not found? Read .relai/notes.md and relai.yaml directly instead — they are plain files, no CLI required.
  2. Before building something another project might already provide, check relai find <capability> (or read a shared .relai/index.json directly, if your group has one) — don't duplicate what already exists elsewhere.
  3. New to the project? Read docs/HANDOFF.md once.
  4. Load anything else only when you need it.

Rules

  • Keep this file under 100 lines.
  • End the commit message of finished work with Closes <task id>: relai tidy (run by the post-merge hook of relai init --hooks) closes the task, refreshes .beads/issues.jsonl and deletes merged branches.
  • Record decisions and pitfalls in .relai/notes.md, not in chat.
  • After creating, updating or closing tasks, run scripts/sync-beads.sh and commit .beads/issues.jsonl: that file is the copy of the task list other people get when they clone.
  • Every bd run rewrites ~3 MiB of its database on disk: never run it on a timer or in a loop.

This repo

  • Go, standard library only unless a dependency is clearly worth it. Single static binary.
  • Before committing: gofmt -l . (must be empty), go vet ./..., go test ./....
  • Everything is written in English (code, docs, commits, issues).
  • Commit messages: conventional style (feat:, fix:, docs:…), no Co-Authored-By or "Generated with" trailers.
  • Design lives in docs/DESIGN.md; changes go through OpenSpec (openspec/), tasks through Beads (bd ready).
  • Keep it easy: every feature must work for one person with only git, no server.

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.