claude-code-kit AGENTS.md

A project instruction file that gives coding agents shared rules for planning, implementation, quality checks, and completion.

In plain words
What is it for?
Use it to define agent behavior, require plans and verification before completion, and document limits of the rules copied from claude-code-kit.
Why use it?
It gives different coding agents the same project workflow and records which rules are automatically generated versus project-specific.

Instructions file for CodexOpenCode

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/this-hw/claude-code-kit/agents-md
Clone the repo
git clone --depth 1 https://github.com/This-HW/claude-code-kit

Made for: Codex, OpenCode.

Per session 6,397 This file is loaded in full into every session.
When invoked 6,397 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.06397 $0.06397
Opus 5 $0.03198 $0.03198
Sonnet 5 $0.01279 $0.01279
Haiku 4.5 $0.00640 $0.00640

Measured yesterday against content hash 873795c2db77, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

claude-code-kit AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 485 lines

How it starts

The opening of the file, as written. The whole thing — 485 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

이 파일의 cck: 마커 블록은 자동 생성된다. 재생성: ./scripts/export-harness.sh (플러그인 사용자는 /harness-export 스킬 참조) 마커 블록 밖의 내용은 생성기가 건드리지 않는다 — 프로젝트 고유 규약을 자유롭게 적어라.

claude-code-kit — 하네스 중립 규범

이 절은 자동 생성된다. 위아래의 cck 주석 마커 사이는 재생성 시 통째로 교체되고, 그 밖은 생성기가 건드리지 않는다. 갱신은 /harness-export 스킬(또는 kit 레포에서 ./scripts/export-harness.sh). 손으로 고치면 드리프트 검사가 막는다.

이 절은 claude-code-kit의 규범을 원문 그대로 옮긴 것이다. Claude Code·Codex·OpenCode·Copilot·Pi·Hermes 등 이 파일을 읽는 모든 에이전트에 동일하게 적용된다.

워크플로 체인

brainstorming  →  plan-task  →  auto-dev
   (설계·스펙)     (구조화 계획)   (구현 + 검증)

각 단계는 앞 단계의 산출물 없이 시작하지 않는다. 완료 선언 전에는 프로젝트의 검증 명령을 실제로 실행하고 그 출력을 근거로 삼는다 (아래 definition-of-done).

이식된 룰

이식 사유
rules/code-quality 코드 품질 규범 — 호스트 무관
rules/definition-of-done 완료 판정 규율 — 호스트 무관
rules/feedback-loop 결함 학습 루프 — 호스트 무관
rules/loop-engineering 루프/재시도 규율 — 호스트 무관
rules/planning-check 계획 전 확인 규율 — 호스트 무관
rules/planning-protocol 계획 수립 프로토콜 — 호스트 무관
rules/ssot 단일 진실 원천 규범 — 호스트 무관
rules/tool-usage-priority 도구 선택 우선순위 — 개념 수준에서 호스트 무관

이 파일이 이식하지 못하는 것 (정직한 한계)

영역 이유
훅 (protect-sensitive · stop-validator · auto-format) Claude Code 훅 런타임 전용 — 다른 하네스에는 실행 지점이 없다
서브에이전트 정의 (33종) Claude Code 서브에이전트 규격 전용
룰 본문의 kit-레포 전용 명령 (scripts/verify-done.sh 등) "요약 금지 / 원문 그대로" 정책의 대가 — 각 룰이 "이 레포에선"으로 한정하고 있으니, 당신 프로젝트의 해당 명령으로 읽어라
rules/agent-delegation-chain Claude Code 서브에이전트 호출·수명주기 규격에 종속
rules/agent-system Claude Code 서브에이전트 정의 규격에 종속
rules/mcp-usage Claude Code의 MCP 도구 allowlist 규격에 종속
rules/parallel-worktree isolation: worktree 프론트매터(네이티브 프리미티브)에 종속
rules/task-resume Claude Code Task 도구 수명주기에 종속

즉 다른 하네스에서 이 규범은 규율 문서로 동작하지 강제 장치로 동작하지 않는다. 강제가 필요하면 그 하네스의 네이티브 수단(pre-commit 훅, CI)에 같은 검사를 걸어라.

Read the full file on GitHub · 485 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 485 lines · 6,397 tokens per session scan A 873795c2db77

Subscribe to this mod's changes

claude-code-kit AGENTS.md is an instructions file published in the GitHub repository This-HW/claude-code-kit (4 stars, last pushed 4d ago), licensed MIT. It adds 6,397 tokens to every session, about $0.0320 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

msp-claude-plugins AGENTS.md

Instructions for WYRE-AI/msp-claude-plugins, covering agents.md — guidance for ai agents, what this repo is, directory structure, writing skills and skill title.

WYRE-AI/msp-claude-plugins · 1,066 tokens

agent-plugins-skills copilot-instructions.md

Instructions for richfrem/agent-plugins-skills, covering copilot instructions for agent-plugins-skills, 1. think before coding, 2. simplicity first, 3. surgical changes and 4. goal-driven execution.

richfrem/agent-plugins-skills · 6,436 tokens

agents-plugins AGENTS.md

Instructions for CitiesSkylinesModding/agents-plugins, covering agents.md, project overview, tech stack, repository structure and commands.

CitiesSkylinesModding/agents-plugins · 3,456 tokens

claude-legal CLAUDE.md

Claude Code instructions for codearranger/claude-legal, covering claude.md, what this repo is, big-picture architecture, plugins at a glance and reference corpora at a glance.

codearranger/claude-legal · 13,184 tokens

agent-plugins-skills AGENTS.md

Instructions for richfrem/agent-plugins-skills, covering agents.md, 1. think before coding, 2. simplicity first, 3. surgical changes and 4. goal-driven execution.

richfrem/agent-plugins-skills · 6,397 tokens

the-unofficial-swift-programming-language-skill CLAUDE.md

Instructions for Shada01245/the-unofficial-swift-programming-language-skill, covering claude.md, repository purpose, build commands, generate the skill package and build options.

Shada01245/the-unofficial-swift-programming-language-skill · 1,017 tokens