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.
npx agentmods add instructions/sunrnehui/agent-harness/agents-mdgit clone --depth 1 https://github.com/SUNRNEHUI/agent-harnessWrote 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.
[](https://agentmods.dev/instructions/sunrnehui/agent-harness/agents-md)<a href="https://agentmods.dev/instructions/sunrnehui/agent-harness/agents-md"><img src="https://agentmods.dev/badge/instructions/sunrnehui/agent-harness/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01481 | $0.01481 |
| Opus 5 | $0.00740 | $0.00740 |
| Sonnet 5 | $0.00296 | $0.00296 |
| Haiku 4.5 | $0.00148 | $0.00148 |
Grade A, and why
agent-harness 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
本文件是 agent-harness 项目的项目级规则。它覆盖用户级通用规则中与本项目冲突的部分。
项目定位
本仓库维护一个运行时中立、Plan-native 的可靠性 skill。它复用 Codex、Claude Code、Grok 等运行时的原生规划，只在任务需要跨会话、跨模型或强审计时增加持久状态。
核心原则：
- 默认选择能可靠完成任务的最轻模式：Native、Portable、Audited。
- Native 使用运行时原生 Plan，不写 harness 文件，也不复制计划。
- Portable 只保存
contract.json、events.jsonl、capsule.md；contract.json是唯一可变真相。 - Audited 才启用 typed receipts、Production State Witness、受保护 TDD chronology、evaluator 和更强 fencing。
- 并行是执行选择，不是模式；只有责任边界独立且收益高于协调成本时才使用 worker。
- Portable 核心只保存可观察事实，不保存 chain-of-thought、完整聊天、provider reasoning、secret 或 in-flight 外部副作用。
- 保持 legacy
handoff-v1Full artifact 可恢复，不静默迁移或破坏旧格式。 - provider/model slug 只能存在于 runtime profile 或 adapter，不进入 Portable contract。
主要文件
skills/agent-harness/：唯一、完整、可直接安装的 runtime Skill 包。skills/agent-harness/SKILL.md：Skill 入口协议和触发说明。skills/agent-harness/VERSION：当前发布版本的单一来源。README.md/README.zh-CN.md：公开说明，必须保持英文和中文同步。skills/agent-harness/references/：运行时按需加载的协议材料。skills/agent-harness/templates/：运行时需要的 artifact 与 worker 模板。skills/agent-harness/scripts/：控制器、validator、status、TDD 和 witness 运行时脚本。tests/：行为、拒绝边界和开放包结构契约测试；不得放进 runtime Skill 包。scripts/package_skill.py：校验、复制并比较权威skills/包，不维护第二份文件白名单。scripts/sync_version.py：从 Skill 包内的VERSION同步公开版本引用。docs/legacy/：不进入 runtime 的历史 prompt、旧协议说明和开发工具。
修改规则
- 修改 skill 行为时，优先改行为测试，再改 Skill 包内的
SKILL.md、references/、模板和脚本。 - 修改公开行为或版本时，必须同步
README.md和README.zh-CN.md。 - 修改当前版本时，先改 Skill 包内的
VERSION，再运行python3 scripts/sync_version.py --fix --date YYYY-MM-DD。 skills/agent-harness/目录本身就是 runtime 权威边界；新增 runtime 文件不需要再登记白名单。- repository-only 测试、fixture、历史文档和维护工具不得进入 Skill 包。
- 不要把
workspace/、.harness/、缓存、session 日志、私有配置或生成 artifact 加入 runtime 包。 - 不要把本地安装目录
~/.codex/skills/agent-harness当成源码。源码以本仓库为准。 README.md与README.zh-CN.md的标题结构必须一致。- Skill 包内的
SKILL.md默认不超过 750 words，agents/openai.yaml不超过 120 words。 - 保持 diff 小而聚焦，不做无关重排、格式化或重命名。
TDD 和验证
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.
- 4d ago First seen · 117 lines · 1,481 tokens per session scan A 8dead16216ee
agent-harness AGENTS.md is an instructions file published in the GitHub repository SUNRNEHUI/agent-harness (2 stars, last pushed 6d ago), licensed MIT. It adds 1,481 tokens to every session, about $0.0074 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.
Other instructions, from other repositories
Traceplane AGENTS.md
Instructions for jaxxchen003/Traceplane, covering agents.md, 1. 实现顺序, 2. 绝对约束, 3. 保底字段 and 4. 模块边界.
Traceplane CLAUDE.md
Instructions for jaxxchen003/Traceplane: This repository exposes an MCP server for Traceplane.
solo AGENTS.md
AGENTS.md instructions for solo-agent/solo, covering project testing rules and project service lifecycle rules.
research-os AGENTS.md
Instructions for lxinfei5/research-os, covering researchos — constitution, §0 what this is, §1 directory discipline, §2 half-life (innovation 1) — memory design and §3 behavior pillars (innovations 2–5).
AI-System-Design-Consultant CLAUDE.md
Claude Code instructions for deepanshu2711/AI-System-Design-Consultant, covering claude.md, what this is, running it, load-bearing typos — do not "fix" these paths and agent node pattern.
agent-eve CLAUDE.md
Instructions for shehryarsaroya/agent-eve, covering claude.md — agent eve (né the compact, then agent transfer), 0. 🚨 hard rules, 1. what this is, 2. doc map and 3. state of play (2026-07-26).