Safely execute database operations on the project's MySQL DB. Connects to TEST DB directly, renders SQL text for PROD (via DMS / Bytebase / DBA tool — configured per project). Use whenever the user asks to query, inspect, modify, insert, update, delete, or otherwise operate on the project database — in any phrasing like "查测试库", "改一下 sysmenu", "删掉工单", "给我一个正式库的 SQL", etc. ALWAYS go through this skill instead of running mysql / raw SQL directly.
EnglishUse when editing .claude-rules/ruler.yml or .claude-rules/rules/.md files, or when the user asks to add/modify a project rule, write a ruler entry, or debug why a rule did not trigger. Teaches the ruler.yml schema, matching semantics, best practices, and the ruler-engine CLI (lint + dry-run).
English本地归档已完成的需求文档 + 读档查过去相关经验教训。归档:读目录下的 需求.md / -design.md / -plan.md → 抽 5 类 tag(模块/技术/症状/方案/文件)+ 摘要 + 教训 → 倒排索引落盘。读档:给一个新需求,自动查出过去相关归档 top-K 并展示教训。用户任何关于"归档这个需求"、"查过去做过类似的没"、"有没有相关经验/教训"、"读档"、"归档"、"/archive"、"/lookup" 的诉求,必须通过本 skill 完成,不要手动读写 ~/.claude/skills/archive-ops/data/ 下的…
Chinese(内部子 skill,仅由 fullstack-builder 主 skill 通过 Read 调用,不应被顶层直接触发)fullstack-builder 的改动归档子 skill。在 self-review loop 通过后基于 git diff + 需求.md + 代码产出 改动.md(严格按 requirement-schema §2 字段),面向 test-runner…
Chinese全栈开发编排器。当用户意图进入实现阶段时触发——关键词包括"实现需求""开始开发""做这个子需求""把需求做了""前后端一起实现""落地 - "等。读取 需求.md/可行性.md 后编排 superpowers 的 writing-plans / TDD / verification,驱动 self-review 多轮 loop,再调 change-summarizer 产出 改动.md…
Chinese写超大文件(预估或已知 >1000 行)时的防 socket 断开策略。用"骨架 + 哨兵 + 分片 Edit"避免单次 Write 触发 socket connection was closed unexpectedly / API Error / Write failed。任何涉及"写大文件"、"生成完整 HTML/Vue/Python 等长文件"、"一次性生成整个 xxxx.xxx"、"文件超过 1000 行"、"分片写"、"chunked write"、"socket closed"、"connection closed unexpectedly"、"write failed"、"写挂了重试" 的诉求(事前预防 / 事后补救),都必须通过本 skill 完成,不要直接 Write…
Chinese(内部子 skill,仅由 product-designer 主 skill 派发调用,不应被顶层直接触发)拆分大 PRD / 多模块设计稿 / 大需求文档为多个独立子需求,自动分配编号,依赖拓扑排序,用户确认后批量生成 PRD.md +…
Chinese
AGENT SKILLS ATLASIncludes Ruler Authoring and other workflows.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLISupported by the Atlas install prompt. Skill-specific requirements vary.
Pack stats
What’s included
Db Ops
Safely execute database operations on the project's MySQL DB. Connects to TEST DB directly, renders SQL text for PROD (via DMS / Bytebase / DBA tool — configured per project). Use whenever the user asks to query, inspect, modify, insert, update, delete, or otherwise operate on the project database — in any phrasing like "查测试库", "改一下 sysmenu", "删掉工单", "给我一个正式库的 SQL", etc. ALWAYS go through this skill instead of running mysql / raw SQL directly.
Ruler Authoring
Use when editing .claude-rules/ruler.yml or .claude-rules/rules/.md files, or when the user asks to add/modify a project rule, write a ruler entry, or debug why a rule did not trigger. Teaches the ruler.yml schema, matching semantics, best practices, and the ruler-engine CLI (lint + dry-run).
Archive Ops
本地归档已完成的需求文档 + 读档查过去相关经验教训。归档:读目录下的 需求.md / -design.md / -plan.md → 抽 5 类 tag(模块/技术/症状/方案/文件)+ 摘要 + 教训 → 倒排索引落盘。读档:给一个新需求,自动查出过去相关归档 top-K 并展示教训。用户任何关于"归档这个需求"、"查过去做过类似的没"、"有没有相关经验/教训"、"读档"、"归档"、"/archive"、"/lookup" 的诉求,必须通过本 skill 完成,不要手动读写 ~/.claude/skills/archive-ops/data/ 下的 JSON。
Change Summarizer
(内部子 skill,仅由 fullstack-builder 主 skill 通过 Read 调用,不应被顶层直接触发)fullstack-builder 的改动归档子 skill。在 self-review loop 通过后基于 git diff + 需求.md + 代码产出 改动.md(严格按 requirement-schema §2 字段),面向 test-runner 作为唯一权威输入
Fullstack Builder
全栈开发编排器。当用户意图进入实现阶段时触发——关键词包括"实现需求""开始开发""做这个子需求""把需求做了""前后端一起实现""落地 - "等。读取 需求.md/可行性.md 后编排 superpowers 的 writing-plans / TDD / verification,驱动 self-review 多轮 loop,再调 change-summarizer 产出 改动.md 并推进状态
Large File Write
写超大文件(预估或已知 >1000 行)时的防 socket 断开策略。用"骨架 + 哨兵 + 分片 Edit"避免单次 Write 触发 socket connection was closed unexpectedly / API Error / Write failed。任何涉及"写大文件"、"生成完整 HTML/Vue/Python 等长文件"、"一次性生成整个 xxxx.xxx"、"文件超过 1000 行"、"分片写"、"chunked write"、"socket closed"、"connection closed unexpectedly"、"write failed"、"写挂了重试" 的诉求(事前预防 / 事后补救),都必须通过本 skill 完成,不要直接 Write 整个长文件。
Requirement Splitter
(内部子 skill,仅由 product-designer 主 skill 派发调用,不应被顶层直接触发)拆分大 PRD / 多模块设计稿 / 大需求文档为多个独立子需求,自动分配编号,依赖拓扑排序,用户确认后批量生成 PRD.md + 需求.md
Quick start
- Choose your skills. Select individual skills above, or choose the entire pack.
- Copy the install prompt. Use Copy install prompt above the list.
- Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.
Try it with your agent
After installation, start with a request like:
Use the Db Ops skill to help with my task.
Ask me for the context you need before getting started.Requirements & license
- A coding agent that can access your project files and install skills.
- Your agent checks each skill’s instructions, dependencies, and permissions during setup.
- License details are not provided in these listings. Check the author’s terms before using the skills.