Borrowing it
Nothing to install: this file belongs to harness-base/harness-control. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/harness-base/harness-control/main/.agents/skills/hc-add-rule/SKILL.mdgit clone --depth 1 https://github.com/harness-base/harness-controlWrote 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/skills/harness-base/harness-control/hc-add-rule)<a href="https://agentmods.dev/skills/harness-base/harness-control/hc-add-rule"><img src="https://agentmods.dev/badge/skills/harness-base/harness-control/hc-add-rule/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.
<a href="https://agentmods.dev/skills/harness-base/harness-control/hc-add-rule"><img src="https://agentmods.dev/badge/skills/harness-base/harness-control/hc-add-rule.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00132 | $0.03191 |
| Opus 5 | $0.00066 | $0.01596 |
| Sonnet 5 | $0.00026 | $0.00638 |
| Haiku 4.5 | $0.00013 | $0.00319 |
Grade A, and why
hc-add-rule 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.
How it starts
The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
规则的加 / 改 / 删（hc-add-rule）
规则写完没人读、没人拦 = 白写；改了语义 / 删了规则但引用点没跟 = 全仓悬空。本 skill 是规则加 / 改 / 删的统一入口：三步保证它放对地方、会被加载、会被执行；「规则关联对照表」保证一条规则的所有关联点（索引 / 挂钩 / eval 指针 / 引用点 / lessons 销点）加时挂全、改删时跟全。
何时用 / 何时不用
- 用（加）：用户说"以后都要 / 不许 / 必须…"，或你发现一条该固化的规范。
- 用（改）：一条既有规则的语义 / severity / 位置 / 挂钩要变。
- 用（删）：规则过时要收回（连同其全部关联点一起清）。
- 用（晋升）：lessons 里反复同根的坑该升成规则（见「从 lessons 晋升」）。
- 不用：一次性临时提醒（那只记
tasks/lessons.md）。
三步（缺一步，规则就等于白加）
第 1 步：定范围 —— 这规则管谁？
先过 rule-0015 的边界：通用的归控制面（harness 资产不掺具体项目的领域名词 / 业务假设），项目专属的归被管工程（projects/<工程>/）。然后一律入驻就近的 AGENTS.md（ADR-0004），放到"能覆盖其所有目标的最浅 AGENTS.md"：
- harness 全局治理（如"改业务码前先立项"）→ 根
AGENTS.md - 某工程通用（如"这后端时间一律 UTC"）→ 该工程根
projects/<x>/AGENTS.md - 只管某层（如"数据层用 ent、非必要不 raw SQL"）→ 离那层最近的
<dir>/AGENTS.md
第 2 步：写下来 + 登记 —— 让它会被读到
- 在选定的
AGENTS.md加 / 改一条 bullet：一句"必须 / 禁止"（必要时带一行为什么 / 怎么做）。 - 带隐形标记供索引扫描：
<!-- rule: <id> | sev: blocker|warn | eval: <考题号，可空> -->。id 有两种口径，scripts/rules-index.sh全仓扫AGENTS.md、两种都收进 catalog：- 全局数字
rule-00NN：harness 全局规则（根AGENTS.md）。编号取现有最大 +1（查docs/rules/index.yaml），全仓唯一、稳定引用键，被 eval 考题 / ADR 按号引用。 - 工程命名 id
<工程>/<主题>（如backend-service/db-eager-load-fail-fast，中性占位）：被管工程 / 目录级规则。命名空间 = 工程短名，主题 slug 说人话、见名知义。
- 全局数字
- 重生成 catalog：
bash scripts/rules-index.sh（生成docs/rules/index.yaml，禁手改）。加 / 改 / 删都要 regen。 - 自检：
make verify绿（rules 索引无漂移 + 该AGENTS.md有CLAUDE.mdshim）= 已收录、就近可加载。
第 3 步：挂执行 —— 让违反会被发现
按"能不能机器判"分两路：
- 能机器判（某命令 / 字符串、改 A 必须改 B、某路径模式）→
scripts/hook-policy.sh加一条匹配 +scripts/hook-policy.test.sh加正反用例 → 提交 / CI 自动拦。 - 要人判（设计是否合理 / 过度）→
docs/eval/prompts/加一道考题引用规则 id + 登记docs/eval/index.yaml→ 收尾 eval 打分。 - 两者都不便 → 至少完成第 2 步（会被加载），显式标"软约束"。
规则关联对照表（一表两用）
一条规则在仓里有 8 个关联点。**主 agent 加 / 改 / 删时照它当操作清单；hc-rule-reviewer 照它当巡查判据。**机器能查的行 make verify 已兜，reviewer 不重复查。
| # | 关联点 | 要求 | 谁兜底 |
|---|---|---|---|
| 1 | 规则本体 | 就近 AGENTS.md bullet + 隐形标记，id 口径对（全局数字 / 工程命名，见第 2 步）；rule-0015 边界（通用归控制面 / 项目归工程） |
主 agent 第 1/2 步自查 + 用户拍（范围与措辞在引导环节定，reviewer 不复审内容本身） |
| 2 | 索引 | bash scripts/rules-index.sh regen docs/rules/index.yaml（禁手改） |
机检（make verify 的 --check 防漂移） |
| 3 | shim | 所在 AGENTS.md 有同级 CLAUDE.md |
机检（make verify） |
| 4 | 执行挂钩 | 机器判 → hook-policy.sh + 测试；人判 → eval 考题 + 登记；都不便 → 显式标软约束 |
reviewer（挂没挂、挂得对不对） |
| 5 | eval 指针 | 标记里的 eval: NNN 与 docs/eval/ 考题双向对上（标了号考题在、考题引的规则号存在） |
正向存在性 / 登记 = 机检（rules-index --check + verify-eval-materials）；内容真伪（考题真考这条）= reviewer |
| 6 | 引用点（改 / 删要害） | 全仓 grep 该规则 id——skill 正文 / reviewer 判据 / ADR / 模板都可能复述它；改语义 / 删除时引用点全要跟（防"复述点没扫"留悬空） | reviewer |
| 7 | lessons 闭环 | 源自错题晋升的，tasks/lessons.md 对应条标 <!-- opt: rule-00NN --> 销掉（该标记也用于"已被既有规则覆盖、不另升"的条目——两种来路读 lesson 正文分辨，覆盖型不当晋升源核） |
reviewer |
| 8 | 反向引用 | 新规则该不该被现有 skill 引为指针（rule-0015 立完、各 skill 补指针，是先例） | reviewer |
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.
- 9d ago First seen · 99 lines · 132 tokens per session scan A df36120803d0
hc-add-rule is a skill published in the GitHub repository harness-base/harness-control (22 stars, last pushed 1mo ago), licensed MIT. It adds 132 tokens to every session and 3,191 once invoked, about $0.0007 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…