morodomimorodomi/dev-crew/AGENTS.md

dev-crew

AGENTS.md1 starsChanged 26 days agoMITView on GitHub

AI development team as a single Claude Code plugin. Skills available: spec, red, green, refactor, review, commit, orchestrate, diagnose, onboard, sync-skills, skill-maker, cycle-retrospective, codify-insight, learn, evolve, security-scan, attack-report, context-review, generate-e2e, security-audit, php-quality, python-quality, ts-quality, js-quality, flask-quality, flutter-quality, hugo-quality, careful Cycle docs: docs/cycles/YYYYMMDD_HHMM_<topic>.md plan review（Claude 設計レビュー + Codex利用可能時は competitive review）は承認前の spec 内（Step 8）で完了済み。Plan mode を抜けたら /orchestrate を起動する。Edit/Write は直接行わず、必ず /orchestrate に委譲する（プロンプトベースの規律。hook による強制ブロックではない）。

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 31,843 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 repository29 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- attack-reportSkill
- carefulSkill
- codify-insightSkill
- commitSkill
- context-reviewSkill
- cycle-retrospectiveSkill
- diagnoseSkill
- evolveSkill
- flask-qualitySkill
- flutter-qualitySkill
- generate-e2eSkill
- greenSkill
- hugo-qualitySkill
- js-qualitySkill
- learnSkill
- onboardSkill
- orchestrateSkill
- php-qualitySkill
- python-qualitySkill
- redSkill
- refactorSkill
- reviewSkill
- security-auditSkill
- security-scanSkill
- skill-makerSkill
- specSkill
- sync-skillsSkill
- ts-qualitySkill

The file

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

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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

Distinctive:
Read CONSTITUTION.md for principles and docs/workflow.md for development flow
line 7
Very specific:
Check docs/STATUS.md and docs/cycles/ for active work
line 8
Very specific:
If an IN_PROGRESS cycle exists, continue it.
line 9

≈ “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

dev-crew

AI development team as a single Claude Code plugin.

Start Here

Read CONSTITUTION.md for principles and docs/workflow.md for development flow
Check docs/STATUS.md and docs/cycles/ for active work
If an IN_PROGRESS cycle exists, continue it; otherwise start with spec

Tech Stack

Target Languages: PHP, Python, TypeScript, JavaScript, Flutter, Hugo
Security: OWASP Top 10, CWE Top 25

Skills

Skills available: spec, red, green, refactor, review, commit, orchestrate, diagnose, onboard, sync-skills, skill-maker, cycle-retrospective, codify-insight, learn, evolve, security-scan, attack-report, context-review, generate-e2e, security-audit, php-quality, python-quality, ts-quality, js-quality, flask-quality, flutter-quality, hugo-quality, careful

Quick Start

# Run all structure validation tests
bash run-tests.sh

# Run a specific test
bash run-tests.sh tests/test-plugin-structure.sh

TDD Workflow

spec → plan-review → approve → /orchestrate (sync-plan → pre-red-gate → RED → GREEN → REFACTOR → REVIEW → cycle-retrospective → pre-commit-gate → COMMIT)

Cycle docs: docs/cycles/YYYYMMDD_HHMM_<topic>.md

Post-Approve Action

plan review（Claude 設計レビュー + Codex利用可能時は competitive review）は承認前の spec 内（Step 8）で完了済み。Plan mode を抜けたら /orchestrate を起動する。Edit/Write は直接行わず、必ず /orchestrate に委譲する（プロンプトベースの規律。hook による強制ブロックではない）。

Quality Standards

Metric	Target
Test coverage	90%+ (min 80%)
Static analysis	0 errors
Test design	Given/When/Then
SKILL.md	< 100 lines

Key Constraints

Constraint	Rule
SKILL.md	< 100 lines. Detail goes in reference.md (Progressive Disclosure)
Agents	Markdown with YAML frontmatter (name, description, model, tools?). read-only agent は tools で Read/Grep/Glob に限定、writer は tools 省略で全継承。例外: dast-crawler は Playwright MCP のツール名制約（未設定環境で起動拒否）のため tools 省略で全継承（writer ではない）。`memory` を持つ agent は必ず `disallowedTools: Write, Edit` を併記する（memory 併用時は tools 宣言外の Write が有効化されるため）。例外: adversarial-reviewer は不具合の再現のため Bash を持つ。Write/Edit は disallowedTools で外すが、Bash 経由の書き込みは防げないので、orchestrate が `scripts/gates/worktree-fingerprint.sh` の前後比較で作業ツリーの不変を確かめる
Cycle docs	`docs/cycles/YYYYMMDD_HHMM_description.md`
Git	`<type>: <subject>` (feat/fix/docs/refactor/test/chore)
Tests	All changes require passing tests before commit

Structure

dev-crew/
├── agents/          # Agent definitions (flat), security agents included
├── skills/          # Skills (each: SKILL.md + reference.md)
├── scripts/gates/   # Deterministic gate scripts (pre-red, pre-commit)
├── rules/           # Rules by load trigger: always / cycle-scoped / file-scoped
├── tests/           # Shell test scripts
└── docs/            # See docs/README.md
    └── decisions/   # Architecture decisions (ADR)

References

Topic	Location
Constitution (principles)	CONSTITUTION.md
Workflow (dev flow)	docs/workflow.md
Roadmap	ROADMAP.md
Architecture	docs/architecture.md
Skills	Plugin system auto-discovery (SKILL.md description)
Terminology	docs/terminology.md
Status	docs/STATUS.md

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.

Check your setup, free