aod / templates
AGENTS.md0 starsChanged 29 days agoApache-2.0View on GitHub
This repository uses Architect-Orchestrated Development. An architect agent owns decisions, contracts and acceptance tests; implementer agents each work on exactly one task packet inside an isolated worktree slot. Everything written in the repository is in English. <!-- aod:end -->
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 31,439 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 repository17 other files it gives its agents.
- templates/CLAUDE.mdCLAUDE.md
- adrSkill
- closeSkill
- cross-reviewSkill
- decisionSkill
- delegateSkill
- doctorSkill
- exploreSkill
- initSkill
- packetSkill
- preflightSkill
- retroSkill
- session-endSkill
- session-startSkill
- specSkill
- verifySkill
- waveSkill
The file
We split this AGENTS.md into the rules it gives an agent
4 rules. About 1 is common, 2 are distinctive, and 1 is very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈2
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
- could be automatic: a setting could enforce it
- Distinctive: line 32
All acceptance commands of the packet pass.
- Very specific: line 33
Every changed path is inside the packet scope.
- Common: line 41
Add a dependency without an ADR.
- Distinctive: line 42
Put links to AI sessions or chats in commits, PRs or tags.
≈ “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.
Files
{project}
<!-- aod:begin -->How this repository is worked on (AOD)
This repository uses Architect-Orchestrated Development. An architect agent owns decisions, contracts and acceptance tests; implementer agents each work on exactly one task packet inside an isolated worktree slot.
Read before working
| Task type | Read first |
| Any task | this file, .agents/context/invariants.md, your task packet .agents/tasks/T-xxxx.md |
| Implementation | the contracts and ADRs listed in the packet, .agents/rules/ for the language |
| Review | .agents/rules/review-checklist.md, the spec referenced by the packet |
Repository map
| Path | Purpose | Owner |
.agents/ | Project memory: context, decisions, specs, tasks, plans, journal | architect |
schemas/ | Contracts, single source of truth | architect |
tests/acceptance/ | Acceptance tests, written before implementation | architect |
| everything else | Code and unit tests | as assigned by the task packet scope |
Commands
<!-- Fill in: install, format, lint, typecheck, test. Target shell: bash. -->Definition of Done
- All acceptance commands of the packet pass; unit tests of touched packages pass.
- Every changed path is inside the packet
scope. - Docs are updated in the same change when behaviour changes.
Never
- Edit contracts (
schemas/**), acceptance tests (tests/acceptance/**) or accepted ADRs unless you are the architect. - Skip, special-case or weaken a test or tolerance to make it pass. If a test looks wrong, stop and report
blocked. - Push, force-push, rebase, reset --hard, clean, switch branches or delete branches.
- Add a dependency without an ADR. Commit secrets or build artefacts.
- Put links to AI sessions or chats in commits, PRs or tags.
Language
Everything written in the repository is in English.
<!-- aod:end -->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
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.