self-skill-registry

将指定 commit（含）之前的 Git 历史压缩为一个新的根提交，再把之后的提交按顺序 cherry-pick 到新历史上。 触发场景： - 用户说"把某个 hash 之前的历史合并成一个提交" - 压缩/折叠早期噪声提交，只保留近期增量历史 - 发布前整理仓库，生成"一个基线 + 后续增量提交"结构 - 清理历史再对外开源或归档 不适用：含大量 merge commit 的复杂拓扑、多人共享且禁止 force push 的公共分支。

Evaluate Clawdbot skills for quality, reliability, and publish-readiness using a multi-framework rubric (ISO 25010, OpenSSF, Shneiderman, agent-specific heuristics). Use when asked to review, audit, evaluate, score, or assess a skill before publishing, or when checking skill quality. Runs automated structural checks and guides manual assessment across 25 criteria.

Use when implementing OAuth2 login with frontend-backend separation, where frontend redirects to backend OAuth endpoint and receives JWT token via redirect

Package published npm CLI tools into Arch Linux AUR packages with reproducible npm tarball builds. Use when creating or updating PKGBUILD/.SRCINFO for a Node.js or TypeScript command-line package already published to npm, computing sha256 for npm tarballs, inspecting bin entries and postinstall scripts, generating wrapper scripts, or running makepkg verification. Triggers include: \u201c把 npm CLI 打包成 AUR\u201d, \u201c为 npm 包生成 PKGBUILD\u201d, \u201cupdate AUR package for Node.js CLI\u201d, \u201c生成 .SRCINFO\u201d, \u201ccompute sha256 for npm tarball\u201d.

Use when designing, reviewing, or optimizing MySQL database schemas — covering table structure design, fixed audit fields, status field conventions, time field type selection, unique constraints for idempotency, and index design principles with EXPLAIN guidance. Triggers: 在项目中设计、评审或优化 MySQL 表结构，需要统一固定字段、状态字段、时间字段、唯一约束与索引设计规范时。Also relevant for database schema review, SQL table creation, index optimization, database migration design, and multi-table relationship modeling.

当用户要求精简 skill、减少体积、提高命中率或拆分大 skill 时触发。勿用于业务代码、技术文档或插件实现。

Guide users through co-authoring a substantial structured document such as a proposal, spec, RFC, PRD, or decision doc. Trigger only when the user needs a staged writing workflow from context gathering through drafting and reader validation; do not use for small edits, proofreading, translation, or generic Q&A.

Fastapi 项目的全栈开发指南，涵盖 API 路由、SQLAlchemy 异步数据库模型、Repository 模式、依赖注入和测试的完整开发流程。使用场景：添加新的 API 端点和功能模块、创建或修改数据库模型和关系、编写 Repository 查询逻辑、实现完整的 CRUD 操作、编写和维护测试用例、排查数据库和 API 相关问题。适用于需要了解 Fastapi 项目架构和开发规范的所有开发任务。

Use when the user asks about changes between GitHub releases, wants to know what changed from one version to another, or needs a release changelog summary. Works by calling GitHub API via bundled scripts to fetch and summarize releases between two tags. Triggers on: 项目从 v1.0 到 v2.0 有什么变化, 版本之间有哪些更新, what's new in releases, release changelog, list releases between tags.

Use when facing complex, vague, or large-scope tasks that need to be broken down into clear, actionable steps before execution. Triggers include multi-step features, unclear requirements, or goals that cannot be completed in one hour.