aiboarding
AGENTS.md2 starsChanged 3 months agoMITView on GitHub
AIBoarding generates, compresses, audits, and maintains AGENTS.md onboarding guidance for AI coding agents. Portable SKILL.md workflows, Bash hook/tool templates, JSON settings, Markdown docs. Supports Claude Code, Codex, Copilot CLI, Cursor, and OpenCode. Fast/full suite: bash tests/run.sh. Validate plugin before release: claude plugin validate . --strict. Use openspec workflows for planned changes. skills/ owns lifecycle reasoning. templates/ owns installed hooks, tools, state defaults, and settings. tests/ verifies deterministic behavior. docs/VERIFICATION.md holds manual runtime protocols. Flow: skills synthesize guidance and install template…
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 32,671 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 repository26 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- cutting-a-releaseSkill
- more-with-lessSkill
- openspec-apply-changeSkill
- openspec-archive-changeSkill
- openspec-continue-changeSkill
- openspec-exploreSkill
- openspec-ff-changeSkill
- openspec-new-changeSkill
- openspec-onboardSkill
- openspec-proposeSkill
- openspec-sync-specsSkill
- openspec-update-changeSkill
- openspec-verify-changeSkill
- cutting-a-releaseSkill
- openspec-apply-changeSkill
- openspec-archive-changeSkill
- openspec-continue-changeSkill
- openspec-exploreSkill
- openspec-ff-changeSkill
- openspec-new-changeSkill
- openspec-onboardSkill
- openspec-proposeSkill
- openspec-sync-specsSkill
- openspec-update-changeSkill
- openspec-verify-changeSkill
The file
We split this AGENTS.md into the rules it gives an agent
5 rules. About 0 are common, 3 are distinctive, and 2 are 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.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
Use openspec workflows for planned changes.
- Distinctive: line 29
Keep hooks deterministic, small, and silent by default.
- Distinctive: line 33
Do not duplicate canonical guidance in CLAUDE.md.
- Very specific: line 37
Run bash tests/run.sh. For self-host changes, run bash tests/self-host/test-managed-repo.sh, .aiboarding/tools/check-size-budget AGENTS.md, …
- Distinctive: line 41
Ask before changing public lifecycle semantics, plugin packaging, compatibility, or generated guidance content requiring product judgment.
≈ “Common” is an estimate: at least 1 in 5 of the 41,581 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 Purpose
AIBoarding generates, compresses, audits, and maintains AGENTS.md onboarding guidance for AI coding agents.
Stack and Runtime
Portable SKILL.md workflows, Bash hook/tool templates, JSON settings, Markdown docs. Supports Claude Code, Codex, Copilot CLI, Cursor, and OpenCode.
Build, Test, Run
Fast/full suite: bash tests/run.sh.
Validate plugin before release: claude plugin validate . --strict.
Use openspec workflows for planned changes.
Architecture Map
skills/ owns lifecycle reasoning. templates/ owns installed hooks, tools, state defaults, and settings. tests/ verifies deterministic behavior. docs/VERIFICATION.md holds manual runtime protocols.
Flow: skills synthesize guidance and install template copies; hooks detect drift; update-agent-onboarding patches affected guidance or advances only .aiboarding/state.json.
Domain Model
Canonical guidance: AGENTS.md. Claude adapter: CLAUDE.md importing @AGENTS.md. Operational state: .aiboarding/state.json. Configuration: .aiboarding/config.json. Sync pointer remains outside instruction files to prevent drift loops.
Agent Guardrails
Keep hooks deterministic, small, and silent by default. Keep reasoning in skills, not hooks or tests. Do not add self-host-only lifecycle code. Preserve fixtures and public lifecycle behavior. Installed .aiboarding/hooks/ and .aiboarding/tools/ must byte-match templates/.
Known Failure Modes
Do not duplicate canonical guidance in CLAUDE.md. Do not modify instruction files during no-op drift updates. Do not ignore README.md for this repository: it is authoritative onboarding input. Do not replace manual reasoning protocols with inaccurate Bash simulations.
Verification Before Completion
Run bash tests/run.sh. For self-host changes, run bash tests/self-host/test-managed-repo.sh, .aiboarding/tools/check-size-budget AGENTS.md, preservation checks, and audit workflow. Run claude plugin validate . --strict before release work.
Escalation - Ask the User When
Ask before changing public lifecycle semantics, plugin packaging, compatibility, or generated guidance content requiring product judgment. Ask before destructive Git operations or when self-host setup would alter user-owned settings content.
More With Less engineering policy
For non-trivial project work, load and follow $more-with-less before choosing the implementation approach. Build the minimum sufficient solution while preserving correctness, security, data integrity, accessibility, operability, acceptance criteria, and rollback safeguards. Understand relevant behavior before simplifying; prefer deletion/reuse, native mechanisms, installed dependencies, deterministic checks, one agent by default, and proportional specification/verification. Before completion, run the authoritative project check, report the evidence obtained, and inspect the diff for unrelated or unnecessary changes.
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.

