harness-lab: Instructions file for Claude Code

CLAUDE.md

harness-lab CLAUDE.md is an instructions file for Claude Code from Joe-rq/harness-lab. It costs 2,685 tokens per session, scanned A, original, MIT.

A set of project instructions for harness-lab, including rules for starting work, tracking requirements, and deciding when a change needs a formal request.

In plain words
What is it for?
Use it when working in the harness-lab repository to check active requirements, resume progress, split large changes into smaller requests, and follow the project's implementation process.
Why use it?
It helps prevent coding from starting without the right project context or an appropriately sized plan.

Instructions file for Claude Code

Written for Claude Code: PreToolUse hook event. Also seen: mentions CLAUDE.md.

This is Joe-rq/harness-lab's own configuration. It tells Claude Code how to work on harness-lab itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything harness-lab configures →

Reuse

Borrowing it

Nothing to install: this file belongs to Joe-rq/harness-lab. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Joe-rq/harness-lab/master/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/Joe-rq/harness-lab

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for harness-lab CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/joe-rq/harness-lab/claude-md/github.svg)](https://agentmods.dev/instructions/joe-rq/harness-lab/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/joe-rq/harness-lab/claude-md"><img src="https://agentmods.dev/badge/instructions/joe-rq/harness-lab/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for harness-lab CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/joe-rq/harness-lab/claude-md"><img src="https://agentmods.dev/badge/instructions/joe-rq/harness-lab/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 2,685 This file is loaded in full into every session.
When invoked 2,685 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.02685 $0.02685
Opus 5 $0.01342 $0.01342
Sonnet 5 $0.00537 $0.00537
Haiku 4.5 $0.00268 $0.00268

Measured 9d ago against content hash 927a1e6fd138, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

harness-lab CLAUDE.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 9d ago.

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.

CLAUDE.md · 256 lines

How it starts

The opening of the file, as written. The whole thing — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

⚠️ 会话启动协议

在回复用户的第一个消息之前,必须执行以下步骤:

  1. 读取 requirements/INDEX.md 确认当前活跃 REQ
  2. 读取 .claude/progress.txt 恢复上下文
  3. 在回复开头声明当前状态

状态声明格式:

📊 当前状态: [REQ编号] / [阶段] / [下一步]

如果用户没有指定任务,不要主动开始实现。先确认:

  • 当前是否有活跃 REQ?
  • 用户想要做什么?

🚨 实施前检查点

在写代码之前,必须通过以下检查:

检查 1: 是否需要 REQ?

情况 需要 REQ
涉及 3+ 文件的改动 ✅ 是
新功能开发 ✅ 是
架构/流程变更 ✅ 是
单文件小改动(typo、小 bug) ❌ 否
用户明确说"不用 REQ" ❌ 否

检查 2: REQ 颗粒度是否合适?

核心原则:一个 REQ 涉及实体数 ≤ 4

实体定义:修改/新增的文件、涉及的模块、关联的概念。

涉及文件数 判断
1-2 ❌ 不需要 REQ
3-4 ✅ 需要 1 个 REQ
5-8 ⚠️ 需要 2 个 REQ(按功能边界拆分)
9+ ❌ 必须拆分为 3+ 个 REQ

颗粒度自检清单(创建 REQ 时必填):

1. [ ] 目标数 ≤ 4?
2. [ ] 涉及文件数 ≤ 4?
3. [ ] 涉及模块/目录 ≤ 4?
4. [ ] 能否用一句话描述"解决了什么问题"?
5. [ ] 如果失败,能否干净回滚?

任何一项回答"否"→ 考虑拆分。

检查 3: REQ 是否存在?

如果需要 REQ:
  ├── 有活跃 REQ → 继续实施
  └── 无活跃 REQ → 先创建 REQ,再实施

检查 4: 计划来源是什么?

重要: 用户给出的"实施计划"不等于 REQ!

  • 用户给计划 → 询问是否需要创建 REQ → 创建 REQ → 实施
  • 用户说"直接做"且符合小改动标准 → 可以跳过 REQ

违规示例(禁止)

❌ 用户提供详细计划 → 直接实施 → 结束
✅ 用户提供详细计划 → 创建 REQ → 实施 → 生成报告

强制执行机制

PreToolUse hook 会强制阻断无活跃 REQ 的文件修改操作:

🚫 REQ ENFORCEMENT: BLOCKED
No active REQ found. File modifications require a REQ for:
  - 3+ file changes
  - New feature development
  - Architecture/flow changes

Hook 也会检测 REQ 是否有实际内容:

  • 如果 REQ 仍包含模板占位符(如 说明为什么要做这件事。),会被阻断
  • 如果 REQ 状态是 draft,需要先运行 npm run req:start 进入 in-progress 状态

req:create 只会生成骨架文件,不代表这个 REQ 已经可以直接开始实施。 在运行 npm run req:start 之前,必须先把背景、目标、验收标准等关键章节写实;空模板 REQ 会被 req:start 直接拒绝。

REQ 工作流程

# 1. 创建 REQ(自动创建豁免文件)
npm run req:create -- --title "Your feature"

# 2. 填充 REQ 内容(背景、目标、验收标准等)
# 豁免文件已自动创建,可以直接编辑 requirements/in-progress/REQ-xxx-*.md

# 3. 启动 REQ(验证内容完整性,自动删除豁免文件)
npm run req:start -- --id REQ-xxx --phase implementation

# 4. 开始实施

关键点

  • req:create 会自动创建 .claude/.req-exempt 豁免文件
  • req:start 成功后会自动删除豁免文件
  • 如果手动创建豁免文件,应在完成后删除

临时豁免机制

如果需要临时跳过 REQ 检查(如紧急修复、小改动):

Read the full file on GitHub · 256 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. 9d ago First seen · 256 lines · 2,685 tokens per session scan A 927a1e6fd138

Subscribe to this mod's changes

harness-lab CLAUDE.md is an instructions file published in the GitHub repository Joe-rq/harness-lab (20 stars, last pushed 26d ago), licensed MIT. It adds 2,685 tokens to every session, about $0.0134 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 2026-08-30.

Related

Other instructions, from other repositories

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,469 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens