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
rebase
Rebase the current feature branch onto a target branch (default origin/main) and resolve its conflicts.
prototype
Builds throwaway code that answers one design question: a single-file HTML demo to push a state model or business logic through hard cases, or several structurally different UI variants on one route, switchable in the browser. Use when the user wants to check whether logic or a state model feels right, explore what a screen should look like before committing, or settle a wayfinder prototype ticket with something runnable.
pr-comments
Reply in thread to every unresolved PR review comment, once the fixes are pushed.
pr-body
The shape of a pull request description: the smallest visual that makes the change clear, evidence that it works, and a merge-danger call (one-way or two-way door, plus blast radius). Use when writing or rewriting a PR description, and when opening or updating a pull request.
open-pr
Opens a pull request for the current feature branch and sees it through: validate, self-review, simplify, organize commits, push, create the PR, wait for CI, run an independent agent review and fix its findings, then answer any outside (Copilot or human) review. Use when implementation is complete and ready for review, when the user asks to open, raise, or submit a PR, and when the harness-issue skill hands off in PR mode. Also use when asked to work through or consolidate several dependency PRs.
minimal-change
Finds the smallest change that actually solves a coding problem: question whether the code needs to exist, then reach for existing code, the standard library, native platform features, and installed dependencies before writing anything new. Use before adding a dependency, an abstraction, a config option, or a new module; when simplifying a diff before review; and when the user asks for the simplest solution or complains about over-engineering. Not for trivial edits or for writing prose.
improve-codebase-architecture
Scans a codebase for deepening opportunities (shallow modules that would be easier to test and navigate if their interfaces shrank), presents them as a visual HTML report, then grills through whichever one the user picks. Use when the user asks for an architecture review, wants to find refactoring opportunities, or says the codebase is getting hard to change.
implement
Build a ticket or small spec in this session, test-first, through to an open PR.
diagnosing-bugs
A disciplined diagnosis loop for hard bugs and performance regressions: build a tight feedback loop that goes red on this bug, minimise, hypothesise, instrument, fix at the root cause with a regression test, and clean up. Use when the user says "debug" or "diagnose", or reports something broken, throwing, failing, flaky, or slow.
commit
Creates conventional commits with quality gates — stages intentionally, checks for uv.lock drift, runs the project's verification command, and writes the message. Owns commit mechanics for the PR workflow skills. Use when the user asks to commit, stage, or write a commit message, and when another skill (open-pr, review-pr) needs the commit mechanics or the uv.lock drift check.
codebase-design
Shared vocabulary and principles for designing deep modules: a lot of behaviour behind a small interface, at a clean seam, testable through that interface. Use when designing or reshaping a module's interface, deciding where a seam goes, making code more testable or easier for agents to navigate, and when another skill needs the deep-module vocabulary.
claude-to-im
Bridge THIS Claude Code or Codex session to Telegram, Discord, Feishu/Lark, QQ, or WeChat so the user can chat with Claude from their phone. Use for: setting up, starting, stopping, or diagnosing the claude-to-im bridge daemon; forwarding Claude replies to a messaging app; any phrase like "claude-to-im", "bridge", "消息推送", "消息转发", "桥接", "连上飞书", "手机上看claude", "启动后台服务", "诊断", "查看日志", "配置". Subcommands: setup, start, stop, status, logs, reconfigure, doctor. Do NOT use for: building standalone bots, webhook integrations, or coding with IM platform SDKs — those are regular programming tasks.
code-reviewer
Reviews a diff across six dimensions — correctness, design, readability, performance, testing, security — plus the repo's documented standards, code smells, and what the diff could delete, and returns findings classified BLOCKING / SUGGESTION / NITPICK. Runs in a forked context so the diff reading stays out of the main conversation. Use when the user asks for a code review, feedback on changes, or a pre-PR check of the working diff.
wizard
Generates an interactive bash wizard that walks a human through steps only they can perform: it opens each page, says what to click and copy, captures the values, and writes them to .env or GitHub secrets. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Not for steps the agent can perform itself.
claude-api
Build apps with the Claude API or Anthropic SDK. TRIGGER when: code imports `anthropic`/`@anthropic-ai/sdk`/`claude_agent_sdk`, or user asks to use Claude API, Anthropic SDKs, or Agent SDK. DO NOT TRIGGER when: code imports `openai`/other AI SDK, general programming, or ML/data-science tasks.
wayfinder
Plan work too big for one session as a map of decision tickets, resolved one at a time.
wait-what
That last message didn't land: re-explain it with the missing context.
ui-review
Audits web UI components for accessibility and UX against WCAG 2.1 AA, reporting each violation as `[CRITICAL|IMPORTANT|MINOR] — file:line` with a specific fix. Use when the user asks about accessibility, a11y, WCAG, screen readers, keyboard navigation, or color contrast, or wants a UX pass over components.