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
slop-check
Use when the user asks if text or an image is AI-generated, to clean AI artifacts or invisible characters, or to score readability (Flesch-Kincaid) via Slop or Not Pro on-device tools (macOS only) ; do NOT use for rewriting or humanizing text (use agentic-humanizer or humanizer) ; returns AI-detection verdict, cleaned text, or readability grade.
slides
Create and edit presentation slide decks (`.pptx`) with PptxGenJS, layout helpers, and render/validation utilities. Use when building a new PowerPoint deck, recreating slides from screenshots/PDFs/reference decks, modifying slide content while preserving editable output, adding charts/diagrams/visuals, or diagnosing layout issues like overflow, overlaps, and font substitution.
skill-tree-reconcile
Use when asked to sync skills, reconcile skill trees, deploy skills to live, choose which tree wins, compare live vs vault vs repo, run sync-from-live, or repair drift - classifies differences and applies a human-reviewed direction.
skill-stocktake
Use when auditing Claude skills and commands for quality. Supports Quick Scan (changed skills only) and Full Stocktake modes with sequential subagent batch evaluation.
skill-scout
Search local, marketplace, GitHub and web sources before creating a new skill. Use when the user wants to find, fork or build one. NOT for authoring it (use writing-skills).
simplest-probe-first
Use BEFORE committing to any expensive or multi-step approach (vision/OCR pipeline, multi-agent fan-out, live RE, big rewrite, long orchestration). Run the cheapest deterministic diagnostic FIRST to check whether a simpler path works. Prevents the waste where hours go into a heavy pipeline a one-command check would have ruled out (e.g. a 10-hour vision run on a PDF with no text layer).
silence-trim
Use when removing long silences / dead air / filler-pauses from a TALKING video to produce a tight, jump-cut style edit (podcast, vlog, lecture, screen recording, talking-head short). Detects silent spans with ffmpeg `silencedetect` (noise floor in dB + a minimum silence duration), inverts them to KEEP spans leaving a small pad on each edge so cuts don't sound clipped, then re-cuts frame-accurately with a `trim`/`atrim` + `concat` filter_complex — and re-times caption cues onto the compacted timeline so subtitles don't drift late after every removed gap. Do NOT use for: removing background noise while keeping length (that's denoise/filtering, not trimming); music tracks or ambient/B-roll with no speech (everything reads as "silence" and you gut the clip — or as "loud" and nothing cuts); VAD-grade speech/non-speech where the noise floor is high (silencedetect is amplitude-only, not a voice classifier); or audio-event scoring (laughter/applause salience — that's the PANNs audio_saliency path, a different module). Source of truth: sidecar/media_studio/features/silencetrim.py.
sentry
Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry API; perform read-only queries with the bundled script and require `SENTRY_AUTH_TOKEN`.
semgrep
Run Semgrep static analysis scans and create custom detection rules. Use when asked to scan code with Semgrep, find security vulnerabilities, write custom YAML rules, or detect specific bug patterns. IMPORTANT: Also use when users ask to 'scan for bugs', 'check code quality', 'find vulnerabilities', or 'audit this code' - even without naming Semgrep. Pattern-based scanning across 30+ languages.
seed-golden-journey
Use when an app needs a keystone acceptance test that proves the real product works end to end, not just that units pass or coverage is high. Scaffolds a HELD-OUT golden-journey test that launches the REAL app, drives the ONE primary user journey, and asserts a REAL artifact on disk (ffprobe media file, a downloaded file) rather than element-present. Parameterized for Electron, web, and CLI.
security-threat-model
Repository-grounded threat modeling that enumerates trust boundaries, assets, attacker capabilities, abuse paths, and mitigations, and writes a concise Markdown threat model. Trigger only when the user explicitly asks to threat model a codebase or path, enumerate threats/abuse paths, or perform AppSec threat modeling. Not for general architecture summaries or code review.
security-scan
Scan your Claude Code configuration (.claude/ directory) for security vulnerabilities, misconfigurations, and injection risks using AgentShield. Checks CLAUDE.md, settings.json, MCP servers, hooks, and agent definitions.
security-review
Use when adding authentication, handling user input, secrets, API endpoints, or payment/sensitive features -- gives a security checklist and patterns. NOT for scanning your .claude config (use security-scan).
security-ownership-map
Analyze git repositories to build a security ownership topology (people-to-file), compute bus factor and sensitive-code ownership, and export CSV/JSON for graph databases and visualization. Trigger only when the user explicitly wants a security-oriented ownership or bus-factor analysis grounded in git history. Not for general maintainer lists or non-security ownership questions.
security-bounty-hunter
Hunt for exploitable, bounty-worthy security issues in repositories. Focuses on remotely reachable vulnerabilities that qualify for real reports instead of noisy local-only findings.
security-best-practices
Perform language and framework specific security best-practice reviews and suggest improvements. Trigger only when the user explicitly requests security best practices guidance, a security review/report, or secure-by-default coding help. Trigger only for supported languages (python, javascript/typescript, go). Do not trigger for general code review, debugging, or non-security tasks.
search-first
Research-before-coding workflow. Search for existing tools, libraries, and patterns before writing custom code. Invokes the researcher agent.
screenshot
Use when the user asks for a desktop, system or OS-level screenshot (full screen, a window, or a pixel region), or when tool-specific capture is unavailable. NOT for a web page or app UI (use playwright or ui-audit).