GITHUB AGENT SKILLS / 技能收录
给你的 AI 助手
装上专业技能
我们从 GitHub 持续收录、审核并分类优质的 Agent Skills,按职业与使用场景整理,复制一行命令即可安装到 Claude Code、Cursor 等智能体。
按职业浏览 Agent Skills
为不同岗位挑选的高频技能
按使用场景浏览 Agent Skills
从你要完成的任务出发
按创作者浏览 Agent Skills
按 GitHub Stars 排序的优秀创作者
Microsoft
@microsoftopenclaw
@openclawAnthropic
@anthropicsOpenAI
@openaiThe Apache Software Foundation
@apacheGitHub
@githubAutoGPT
@Significant-GravitasGarry Tan
@garrytan全部 Skills 482556
git-workflow
Squad branching model: dev-first workflow with insiders preview channel
distributed-mesh
How to coordinate with squads on different machines using git as transport
client-compatibility
Platform detection and adaptive spawning for CLI vs VS Code vs other surfaces
cli-wiring
暂无简介
ci-validation-gates
Defensive CI/CD patterns: semver validation, token checks, retry logic, draft detection — earned from v0.8.22
architectural-review
How to review PRs for architectural quality — module boundaries, dependency direction, export surface, pattern consistency
architectural-proposals
How to write comprehensive architectural proposals that drive alignment before code is written
agent-conduct
Shared hard rules enforced across all squad agents
agent-collaboration
Standard collaboration patterns for all squad agents — worktree awareness, decisions, cross-agent communication
openspec-propose
Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
openspec-explore
Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
openspec-archive-change
Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
openspec-apply-change
Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
wireframe
Produce low-fidelity wireframes or storyboards to explore a flow, layout, or idea across multiple options before committing to hi-fi design. Use when the user wants to 'explore options', 'sketch something out', 'see a few directions', or when the problem is open enough that hi-fi work would be wasted.
video-test
Record a video test of any feature or user flow via Playwright. Use when the user says "record video test", "video test", "record a test", "show me the feature", "record the UI", "capture video", or wants a visual demo of a feature. Accepts a description of what to test as an argument (e.g., `/video-test the brainstorm creation flow`).
risk-assessment
Assess the delivery risk and ambiguity of an issue. Scores three anchored 1-5 risk dimensions — Code Blast Radius, Complexity, Business Blast Radius — into an overall risk level (Low/Medium/High) with safeguards, plus a separate Ambiguity score (how many open questions remain) with the open questions listed. Use this BEFORE work starts, whenever someone wants to know how risky, how big, how complex, or how well-specified an issue is, or asks to "assess risk", "risk assessment", "score this issue", "how risky is this", "what's the blast radius", "how complex is this", "how ambiguous is this", or "how many open questions are there". Accepts a Multica issue ID, a GitHub issue/PR, or a free-text description as the argument.
research
Deep parallel codebase research before planning or implementation. Spawns 4 specialized agents that explore behavior, architecture, coding standards, and test coverage simultaneously, then synthesizes findings into a structured RESEARCH.md document. Use this skill when the user says "research", "investigate", "explore the codebase", "understand the code", "deep dive", or wants thorough analysis of an area before making changes. Also use it as the first step of any complex feature — research first, then plan, then implement.
ponytail
Lazy senior dev mode for any coding task (write, refactor, fix, review): YAGNI, stdlib first, no unrequested abstractions. Not for non-coding requests.