agentskills /agentskills
Specification and documentation for Agent Skills
★ 27,313⑂ 2,035👁 245⑃ 228● 205▤ 0
+24
日增长
+400
周增长
+1,239
月增长
27,313
stars
2,035
forks
245
watchers
228
prs
205
issues
0
releases
README
Agent Skills A standardized way to give AI agents new capabilities and expertise. What are Agent Skills? Agent Skills are a lightweight, open format for extending AI agent capabilities with specialized knowledge and workflows. At its core, a skill is a folder containing a SKILL.md file. This file includes metadata (name and description, at minimum) and instructions that tell an agent how to perform a specific task. Skills can also bundle scripts, reference materials, templates, and other resources. my-skill/ ├── SKILL.md Required: metadata + instructions ├── scripts/ Optional: executable code ├── references/ Optional: documentation ├── assets/ Optional: templates, resources └── ... Any additional files or directories Why Agent Skills? Agents are increasingly capable, but often don't have the context they need to do real work reliably. Skills solve this by packaging procedural knowledge and company-, team-, and user-specific context into portable, version-controlled folders that agents load on demand. This gives agents: - Domain expertise: Capture specialized knowledge — from legal review processes to data analysis pipelines to presentation formatting — as reusable instructions and resources. - Repeatable workflows: Turn multi-step tasks into consistent, auditable procedures. - Cross-product reuse: Build a skill once and use it across any skills-compatible agent. How do Agent Skills work? Agents load skills through progressive disclosure, in three stages: 1. Discovery: At startup, agents load only the name and description of each available skill, just enough to know when it might be relevant. 2. Activation: When a task matches a skill's description, the agent reads the full SKILL.md instructions into context. 3. Execution: The agent follows the instructions, optionally executing bundled code or loading referenced files as needed. Full instructions load only when a task calls for them, so agents can keep many skills on hand with only a small context footprint. Where can I use Agent Skills? Agent Skills are supported by a large number of AI tools and agentic clients — see the Client Showcase to explore some of them! Getting started - Documentation — Guides and tutorials - Specification — Format details - Example Skills — See what's possible - Discord — Share what you're building! Open development The Agent Skills format was originally developed by Anthropic, released as an open standard, and has been adopted by a growing number of agent products. The standard is open to contributions from the broader ecosystem — see CONTRIBUTING.md for how to get involved. License Code in this repository is licensed under Apache 2.0. Documentation is licensed under CC-BY-4.0. See individual directories for details.
Top issues
#15Proposal: include a standard folder where agent skills should be130 comments · 2025-12-19#255Proposal: `.well-known` URI for skill discovery and distribution21 comments · 2026-03-17#105Support new Claude Code skills as commands frontmatter scheme20 comments · 2026-01-24#46support versioning/locking20 comments · 2025-12-24#100Best practice on having a skill use/depend on other skills18 comments · 2026-01-22#114feat: publish `skills-ref` to Pypi17 comments · 2026-02-02#57RFC: Skill Activation Mechanisms - Bridging the Gap Between Discovery and Usage16 comments · 2025-12-31#30Propose having foldername.md as an alternative discovery pattern for skills12 comments · 2025-12-20
Discussions
#668[Tool] skill-check: warns when a copied skill is out of date (Claude Code and Codex), using `metadata.version` / `metadata.source`0 comments · 2026-11-14#664Skill Lens: trace a workflow step to SKILL.md, edit it, and export your own copy0 comments · 2026-11-12#661Skill Artifact Check: an offline check for stale Skill ZIPs and missing helpers1 comments · 2026-11-11#660[Skills] 79 MIT-licensed SKILL.md guides packaged as one plugin marketplace (Claude Code, Cursor, Codex)0 comments · 2026-11-10#659Are Agent Skills the right long-term abstraction for evolving enterprise systems?0 comments · 2026-11-10#652PrintRelay: a CAD handoff skill with a free enclosure fit-check walkthrough1 comments · 2026-11-07#651Proposal: C2PA Content Credentials for Agent Skills via an external manifest URL and signed directory digest1 comments · 2026-11-05#647[Skills] 20 email marketing workflows for campaigns, automation and audience growth0 comments · 2026-11-04
相关报道 (9)
blog.csdn03 AgentSkills 生态体系与跨平台支持全景2026-03-28搜狐AgentSkills:让AI成为你行业的专家利器2026-01-21blog.csdn2025年第5期:Agent skills 智能体构建的下一个前沿2025-12-30QQ NewsAI 智能体的“说明书”转为社区共治:Anthropic 捐赠 Agent Skills 开放标准2026-10-23AIBaseAnthropic 把智能体"说明书"交出去:Agent Skills 标准捐给社区2026-10-23Pasquale PillitteriAnthropic 宣布将 Agent Skills 标准捐赠给 Linux 基金会2026-10-2336氪用Skill炼化员工,这个春天最荒诞的技术神话2026-04-10凤凰网狂揽4万星,换掉OpenClaw太爽了,5美元就能养个AI打工人2026-04-09www.80aj.comAgentSkills:开源AI技能市场,让Cursor与Claude如虎添翼2026-01-23
News & coverage (11)
noteAgent Skills Joins the Linux Foundation: Points to Check Before Using at Work2026-10-31noteAnthropic Donates Agent Skills to AAIF; Progress in Claude's Verification and Provisioning Infrastructure2026-10-23Blockchain NewsAnthropic donates Agent Skills standard to AAIF2026-10-22vc.ruЧто такое Agent Skills: как устроен формат SKILL.md, который Anthropic передала Linux Foundation, и где он работает ...2026-10-21learnUse Agent Skills with GitHub Copilot2026-09-03githubManage agent skills with GitHub CLI2026-04-16OpenToolsAnthropic donates Agent Skills to AAIF, keeping participation open without membership2026-10-25tokenpost.comAnthropic Donates Agent Skills Standard to Linux Foundation AI Group2026-10-23BigGo FinanceAnthropic Donates AI Agent Standard 'Agent Skills' to AAIF, Transferring Name and Domain2026-10-22NVIDIA DeveloperNVIDIA-Verified Agent Skills Provide Capability Governance for AI Agents | NVIDIA Technical Blog2026-05-19