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
nullcost-search
MUST use for developer provider discovery involving free tier, free trial, cheap, pricing, hosting, Node, Next.js, auth, Postgres, database, email API, SaaS stacks, GPU compute, remote MCP endpoints, APIs, or low setup friction. Auto-trigger for normal discovery questions and explicit `/nullcost-search`. Do not use for domains, registrars, TLDs, or exact domain availability.
nullcost-recommend
MUST use for developer provider recommendations involving free tier, free trial, cheap, pricing, hosting, Node, Next.js, auth, Postgres, database, email API, SaaS stacks, GPU compute, remote MCP endpoints, APIs, or low setup friction. Auto-trigger for normal questions like "find me hosting with a free tier" and explicit `/nullcost-recommend`. Do not use for domains, registrars, TLDs, or exact domain availability.
nullcost-catalog
MUST use for developer service/provider/tool questions involving free tier, free trial, cheap, pricing, hosting, Node, Next.js, auth, Postgres, database, email API, SaaS stacks, GPU compute, remote MCP endpoints, APIs, or low setup friction. Call Nullcost before model knowledge or web. Do not use for domains, registrars, TLDs, or exact domain availability.
cuar
Use whenever the user invokes or mentions CUAR, asks whether an unscheduled Codex reset happened, asks whether CUAR reminder tooling is unavailable, or asks whether a CUAR expiration reminder was created. Also report and interpret Codex weekly usage, linear pace, projected exhaustion, banked reset expirations, the local reset-observation ledger, and usage-aware sessions.
tail
Read the latest bounded stdout or stderr from a tracked detached process job. Use to inspect live build progress, benchmark output, test failures, repair diagnostics, or other command output without loading the entire persisted log.
status
Inspect active and recent detached process jobs in a later user-requested turn, retrieve a lightweight activity preview, or wait once when explicitly requested. Use for questions such as "how's the build going?", later user-requested checks of test, inference, data-processing, or repair progress, or diagnosis of a disappeared worker. Never use it to monitor a job from the same turn that launched it or merely because an automatic Goal continuation arrived.
start
Launch an ordinary finite local workload as a durable detached process job, then release the assigning Codex turn instead of monitoring it. Use proactively for downloads, builds, test suites, evaluations, benchmarks, inference/model A/B runs, data jobs, and repairs whose underlying work may exceed 60 seconds or has uncertain duration. The user-visible parent must launch the job directly; never delegate local process execution or monitoring to a subagent.
result
Retrieve output for jobs and automatic CPJ completion notices.
rerun
Launch a finished Codex Process Jobs record again as a new detached job using its validated persisted argv, working directory, and execution mode. Use only when the user explicitly asks or approves rerunning a specific completed, failed, or cancelled build, test, benchmark, inference run, data job, or repair.
cancel
Safely request termination of a tracked detached process group using PID identity validation and a SIGTERM-to-SIGKILL grace period. Use only when the user asks to stop a job; critical repair, migration, firmware, or destructive jobs require explicit risk-aware approval and --force.
rust-reverse-engineering
Use when analyzing a Rust binary without source code, reverse engineering Rust executables or libraries, demangling Rust symbols, recovering likely crate namespaces, tracing panic or unwind paths, identifying FFI boundaries, or mapping behavior from ELF, Mach-O, PE, `.so`, `.dylib`, `.dll`, `.a`, `.rlib`, or object files.
jetformat
Convert Word, Excel, and PowerPoint to PDF, turn PDF back into Word, extract text, validate files, and merge, extract, or render PDFs with the local Jetformat CLI. Use when the user asks to convert Office documents, generate PDF previews, inspect PDFs, or read Office or PDF text. Do not parse DOCX/PPTX/XLSX yourself.
agentgram
Send explicit, user-requested Telegram messages and files from an agent session through the local Agentgram command-line tool.
pdf-monster
Analyze PDF files by converting them into agent-readable text, OCR text, page render images, and extracted embedded images without polluting the working directory. Use when an agent needs to read, summarize, inspect, compare, quote, or reason about PDFs, especially when the foundation model cannot ingest PDFs directly or when tables, figures, scans, screenshots, or layout matter.
design-os-code2flow
Use when working with DESIGN:OS Code2Flow — Local-first AST codebase-to-user-flow mapper and 11-heuristic UX gap auditor across 7 frontend and mobile frameworks.
design-os-3d-blender
Use when working with DESIGN:OS 3D Blender Agent — Parametric 3D mechanical parts generator, headless Blender Python (`bpy`) pipeline, and embedded Three.js CAD showcase engine.
ml-pipeline
MANDATORY whenever a task involves training, fine-tuning, tuning, or evaluating a machine-learning model on data (tabular, time series, text, images — any modality). Enforces a strict 16-step pipeline that starts with inspecting the raw data, gates each phase behind the user's explicit permission, and produces marimo notebooks with matplotlib visuals so the user can see and understand every step. Never jump straight to model training.
skill-skill
Write, upgrade or audit a Claude skill so it actually changes what gets generated, not a pile of notes. Use whenever the user wants to create a skill, turn a conversation, workflow or methodology into a skill, rename it, add research or lessons, asks why a skill isn't taking effect, or wants to publish a skill and get it listed in awesome lists and plugin marketplaces. Covers distilling from real work (corrections, rejections, accepted outputs), forcing functions (required artifacts, gates, checklists, scripts), descriptions that trigger, naming, a static audit with a health score, a with/without effect test, and a publish preflight plus PR triage for listings. Triggers: 写 skill, 改 skill, 沉淀进 skill, 优化 skill, 发布 skill, 提 PR 收录, 仓库机器人报错; スキルを作る, スキル作成; 스킬 만들기, 스킬 개선; tạo skill, viết skill; स्किल बनाना.