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 skills/xcodethink/open-claude-code-skills/16npx skills add xcodethink/open-claude-code-skills --skill 16git clone --depth 1 https://github.com/xcodethink/open-claude-code-skillsWrote 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/xcodethink/open-claude-code-skills/16)<a href="https://agentmods.dev/skills/xcodethink/open-claude-code-skills/16"><img src="https://agentmods.dev/badge/skills/xcodethink/open-claude-code-skills/16.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 | $0.00238 | $0.03099 |
| Opus 5 | $0.00119 | $0.01550 |
| Sonnet 5 | $0.00048 | $0.00620 |
| Haiku 4.5 | $0.00024 | $0.00310 |
Grade A, and why
16-开源发布 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 — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
开源发布
核心原则：公开仓库 / npm tarball / GitHub Release / git tag 中，零跨项目泄露、零内部规划、零个人路径、零自家产品硬编码。
历史教训（一次开源发布事故）：将一个开源 npm 包发布到 GitHub + npm 时，AI 反复声称"已检查、无敏感信息"，但实际上把内部规划文档（PLAN.md / docs/research/ / docs/SLO.md）、营销文案（X 推文 / show-hn.md / launch-post / 博主外联清单）、跨项目引用（sibling project URL / 品牌名 / 文件路径如
<internal-repo>/...、~/Developer/...）、自家产品硬编码示例（scenarios 全部用某产品 URL、personas 围绕某产品评测剧本、config 以某产品命名、CI workflow 默认值是某产品名）全部泄露到公开仓库 + npm。用户被迫多轮追加 cleanup（v1.1.2 → v1.1.3 → v1.1.4），过程中 git filter-repo 重写历史 + force-push 5 个 tag + npm deprecate 5 个旧版本。用户对"我自己审过了，没问题"这种声明零容忍。
本目录是上述教训的工程化沉淀。任何项目准备开源前，必须 100% 走完。
触发场景
| 场景 | 必读子文件 |
|---|---|
| 项目第一次决定开源（private → public 转换） | 01-决定开源前必做.md 全套 |
| 任何一次 publish 前（npm publish / PyPI / Docker push / git push tag / 创建 GitHub Release） | 02-Pre-publish 审计清单.md 强制走完 |
| 写 / 维护 bundled scenarios / personas / examples / config / workflow | 03-bundled examples 通用化.md |
| 写 CHANGELOG / Release notes / Release body | 04-CHANGELOG 和 Release 写作.md |
| 发现已经泄露（自查发现 / 被人提醒 / Github 反馈） | 05-事故修复手册.md |
| AI 协助开发任何 OSS 项目时 | 06-AI 协助 OSS 项目铁律.md 必看 |
| 项目存在"私仓 → 公仓 promote"关系（持续同步而非一次性 publish） | 08-开源专属文件夹规范.md 强制建立 opensource/<public-repo>/ 标准目录 |
| 新建仓库前 / 整理混淆命名 / "本地 X 与公开 X 名字一样" | 09-仓库命名约定.md 4 大命名分类（私有 <Handle>-* / staging openskills/* / 公开 collection / 公开独立项目） |
核心铁律（必须内化，不需要每次重读）
铁律 1：默认拒绝
公开仓库 / tarball / Release / tag 中，默认所有内容都是泄露，必须显式批准每一项才能进。package.json#files、.gitignore 都要"白名单 / 黑名单组合"，不能依赖默认行为。
铁律 2：grep 才算审过
任何"我审查了，没有泄露"的说法必须有对应的 grep 命令输出作为证据。空口声明 = 没审。
铁律 2b：词表匹配不是审计 —— 必须「枚举 + 判定」
实战教训：一份 9 个词的词表扫出 "0 命中"，据此汇报"无泄露 ✅"四次；实际仓里躺着 5 处内部名，全都不在那 9 个词里。词表只能找到词表里的词 —— 这是它的天花板，填得再全也拦不住你没想到的那个。
- 禁止用"跑了
grep -E '<已知敏感词>'返回 0"当作"不存在"的结论。它只证明了这些词不在，没证明别的不在。 - 正确方法是反过来：枚举仓里全部专有名词形状的 token / URL / 高熵串，逐个判定，跟已审基线比对。这样是 by construction 穷尽的。已固化为
scripts/audit.sh第 5 层。 - 汇报时必须写清覆盖面：「词表扫描 0 命中（仅覆盖 N 个已知词，不是穷尽性结论）」，不能只写 ✅。
What ships with it
14 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- 01-决定开源前必做.md 8.2 KB
- 02-Pre-publish 审计清单.md 10.0 KB
- 03-bundled-examples-通用化.md 10 KB
- 04-CHANGELOG 和 Release 写作.md 8.5 KB
- 05-事故修复手册.md 12 KB
- 06-AI 协助 OSS 项目铁律.md 7.5 KB
- 07-批量项目转-OSS-工作流.md 11 KB
- 08-开源专属文件夹规范.md 11 KB
- 09-仓库命名约定.md 6.2 KB
- 10-开源纪律标准.md 8.4 KB
- agents/openai.yaml 834 B
- leak-keywords.template.txt 2.1 KB
- scripts/pre-publish-audit.sh 8.5 KB runs code
- SKILL.en.md 9.5 KB
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 · 108 lines · 238 tokens per session scan A 37076d8af3de
16-开源发布 is a skill published in the GitHub repository xcodethink/open-claude-code-skills (2 stars, last pushed 29d ago), licensed MIT. It adds 238 tokens to every session and 3,099 once invoked, about $0.0012 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 skills, from other repositories
writing-modpack-changelog
Use when cutting a modpack release. Creates /docs/release-changelog.md if absent; appends a new version section with grouped changes. Triggers - 'cut a release', 'release notes', 'changelog', 'v1.2.3 changes', 'what changed since last version'.
mathodology-dev-test-release
Use when validating the Mathodology skills-only repository or preserving archived knowledge about the former development, testing, deployment, packaging, and release workflows.
release-notes
Draft user-facing release notes from a git log range. Trigger when the user asks for "release notes", "changelog entry", or "what shipped this week" with a git ref range. Output is markdown sections grouped by Features / Fixes / Breaking with PR links.
v1-milestone
Factory line for executing one milestone of the HAR v1.0.0 refactor (epic os-factory/har#225) — plan the wave of parallel subagents, implement each issue in its own HAR slot, ship stacked PRs, run the fixture-e2e milestone gate, and hand off for review. Use when asked to "run the next v1 milestone", "work on v1.0.0"…
releasing-reef
How to cut a reef release — version bump, tag push, and the GitHub Release notes format. Use when the user says "cut a release", "tag vX.Y.Z", "打个 tag 发布", "发布", "bump version", "release notes", or asks "how do I release reef", "what's the notes format", "is it safe to push this tag". Covers the full pipeline pushing…
git-release
Draft release notes from merged PRs, propose a semver bump, and emit a copy-pasteable gh release create command. Use when preparing a tagged GitHub release.