Explore AI Agent Skills & Claude Prompts
Discover open-source agent skills for Claude Code, Codex, ChatGPT, and any tool that uses SKILL.md.
Enter through keywords, occupations, creators, and GitHub sources to see what kinds of skills are emerging across domains.
Use the same catalog through the API
Connect 381,784 public skills to your own search, analytics, or agent workflow with the REST API.
Querying local SQLite index...
teach
by cogni-workInteractive course delivery for learning Claude Cowork and insight-wave plugins. Use this skill whenever the user asks to learn, train, study, or take a course — including "teach me", "start a course", "continue my course", "what courses are available", "how do I use insight-wave", "explain the plugins", "learn how research becomes a report", "learn how trends become solutions", "learn how a portfolio becomes a pitch or website", "learn how a consulting engagement runs end-to-end", "show me how to use Cowork", "train me", "I'm new to insight-wave", "walk me through a workflow", "tour me through research-to-report", "show me an end-to-end pipeline", or any mention of cogni-help, curriculum, or training. Also trigger when someone asks "what can I do with these plugins" or "where do I start" in an insight-wave workspace — they likely need guided learning.
products
by cogni-workDefine and manage the top-level product offerings in the portfolio. Use whenever the user mentions products, product lines, offerings, "what do we sell", product portfolio, product definition, pricing tiers, lifecycle stages, or wants to organize capabilities into named offerings — even if they don't say "product" explicitly.
copy-reader
by cogni-workThis skill should be used when the user wants to review a document from different stakeholder perspectives, simulate how different audiences would read a document, or get multi-perspective feedback before distribution. Common triggers include "review document as stakeholder", "stakeholder review", "reader review", "read as executive", "review from technical perspective", "does this document work for [audience]", "get feedback on this document", "what would [role] think of this", "check if this is ready for stakeholders", or "simulate different readers".
narrative-review
by cogni-workScore and review existing narrative files against story arc quality gates. This skill should be used when the user asks to 'review a narrative', 'score a narrative', 'check narrative quality', 'validate narrative', 'audit narrative', 'grade a narrative', 'evaluate narrative quality', 'narrative scorecard', 'rate my narrative', 'run quality gates on a narrative', or when the narrative-reviewer agent evaluates a generated narrative.
verify-report
by cogni-workVerify claims in a research report against their cited sources using cogni-claims. Auto-detects draft, sources, and prior claim verdicts inside a cogni-research project directory, or runs against any standalone markdown report with inline citations. Use whenever the user says "verify report", "verify claims", "check sources", "fact-check the report", "run claims verification", "check the citations", wants to re-verify after editing a report, or simply says "verify" after a research-report run finishes. Also trigger when a research-report Phase 6 summary recommends it.
wiki-query
by cogni-workAnswer a question by reading the Karpathy-style wiki — never from memory. Claude consults wiki/index.md first, then reads the relevant wiki/<type>/*.md files, synthesizes an answer with [[wikilink]] citations, and optionally files the answer back as a `type: synthesis` page so the knowledge compounds. Use this skill whenever the user says 'query the wiki', 'ask the wiki', 'what do I know about X', 'what does my wiki say about Y', 'wiki query', 'search the wiki for Z', or asks any question after setting up a wiki and expects Claude to reason from it. Also trigger when the user asks 'look up X in the wiki', 'check the wiki for X', or asks a question that clearly lives inside their wiki's domain (e.g. they have an AI-safety wiki and ask about CAI) — offer the wiki as the source of truth.
audit-arcs
by cogni-workAudit cogni-research's local story-arc registry (`references/story-arcs.json`) against cogni-narrative's upstream arc definitions. Detects missing arcs (registry parity), element heading mismatches (EN+DE), proportion drift, and the appearance of new arcs upstream that haven't been mirrored downstream yet. Use whenever the user mentions "audit research arcs", "check research arcs", "narrative drift check", "compare research arcs upstream", "are research arcs up to date", "research arc contract check", "will arc-driven research break on this arc", or any question about whether `cogni-research/references/story-arcs.json` matches the cogni-narrative source of truth — even if they don't say "audit" explicitly. Also use proactively after cogni-narrative version bumps or after editing any per-arc `arc-definition.md` file in cogni-narrative.
portfolio-lineage
by cogni-workTrack source lineage, detect changes in input documents and URLs, and cascade refresh through the feature-proposition-solution chain. Use whenever the user mentions "lineage", "source lineage", "what sources", "what changed", "refresh stale", "source registry", "check sources", "which documents fed", "trace back to source", "show dependencies", "what's affected by", "stale sources", "source drift", "cascade refresh", "where did this come from", "re-upload", or wants to understand or act on the relationship between input sources and portfolio entities.
wiki-refresh
by cogni-workRefresh stale wiki pages with fresh evidence from a completed cogni-research project. Runs wiki-lint internally, matches stale pages to sub-questions via deterministic token overlap (Jaccard), prints a batch plan for one user confirmation, then dispatches wiki-update per matched page with the synthesised refresh content as the new source. Trigger when the user says 'refresh stale pages from research <slug>', 'wiki-refresh against project X', 'update aging pages with the new agent-economy research', 'pull recent findings into stale pages', or 'my wiki is getting stale, use the latest research to refresh it'. Pull-mode only — does not auto-launch new research.
wiki-resume
by cogni-workShow status, activity, health snapshot, and recommended next action for a Karpathy-style wiki — entry count, days since last lint, recent log activity, stale drafts, structural-integrity errors from wiki-health (run automatically every session), and what the user should do next. Use this skill whenever the user says 'resume the wiki', 'wiki status', 'what's in my wiki', 'where was I with the wiki', 'wiki resume', 'show me the wiki overview', or asks 'what should I do next with my wiki'. Also trigger proactively after a wiki-setup or a long gap between invocations to orient the user.
knowledge-report
by cogni-workCompose a research report by reading the bound cogni-wiki knowledge base, then re-deposit the result back into the same wiki — the wiki-roundtrip primitive. Reads .cogni-knowledge/binding.json to resolve the wiki path so the user does not have to. Dispatches cogni-research with report_source=wiki against the bound wiki, runs cycle-guard.py to refuse self-citing loops, then re-deposits via cogni-wiki:wiki-from-research Mode B with the --allow-wiki-source --cycle-guard-cleared opt-in flags. Every deposited page is stamped with derived_from_research:<slug>, and the project is recorded in the binding with the live report_source (wiki or hybrid). Use this skill whenever the user says 'write a report from my <knowledge-slug> knowledge base', 'roundtrip report on X', 'compose from accumulated knowledge', 'synthesise what we know about X from the wiki', 'wiki-report on <topic>'. Phase 2 of the absorption roadmap — proves that knowledge compounds across projects.
knowledge-research
by cogni-workResearch a topic INTO a bound cogni-knowledge base — runs cogni-research on the topic and deposits the findings into the knowledge base's wiki in one prompt. Reads .cogni-knowledge/binding.json to resolve the wiki path so the user does not have to. Every deposited page is stamped with derived_from_research:<slug>, and the project is recorded in the binding's research_projects[] list. Use this skill whenever the user says 'research X into my knowledge base', 'deposit research on X into the eu-ai-act base', 'knowledge research on X', 'add research on X to the knowledge base', 'feed the knowledge base a research run on X'. Knowledge accumulates across runs — the second knowledge-research on a related topic reads what the first deposited.
Browse Agent Skills by Occupation
23 major groups · 867 SOC occupations
Browse by Category
Explore agent skills organized by their primary use case
Explore the agent skills ecosystem by occupation and creator
SkillMD is not just a keyword search box. It is an open map that organizes public skills by occupation, creator, and repository, helping you see which workflows, judgment criteria, and domain habits people are writing for AI agents.
Then follow creators and GitHub repositories back to the source: compare the skills a team maintains, whether the repo is active, and how the README frames the work before you open, install, or reuse anything.
Use it three ways: learn an unfamiliar field by occupation, study how creators organize skills, then use source context to decide what is worth opening or reusing.
01 Map a field
Browse 23 occupation groups and 867 SOC roles to learn what skills exist in adjacent domains and how they break down real work.
02 Follow creators
Use creator and repository pages to inspect maintained skill collections, recent updates, and source context before trusting a result.
03 Search with sources
Search 1.7M+ collected skills, then use occupation tags, creators, and GitHub source context to decide what is worth opening.
Start with the occupation map, then follow creators and repositories back to real code. SkillMD helps explain why a skill is worth opening, not only what it is named.
Standardizing Agent Capabilities with SKILL.md and Model Context Protocol (MCP)
In the rapidly evolving landscape of artificial intelligence, LLM agents (Large Language Model agents) have transitioned from simple text predictors to autonomous problem solvers. To orchestrate complex, multi-step agentic workflows, developers require a standardized format to specify agent capabilities, prompt instructions, system rules, and database bindings. This is where SKILL.md and the Model Context Protocol (MCP) have emerged as standard developer paradigms. SkillMD serves as the central directory for indexing, exploring, and sharing these critical agent configurations.
Our open-source registry currently tracks over 1.7 million collected SKILL.md configurations and system prompts. By compiling agent configurations from active developers on GitHub, we bridge the gap between prompt engineering research and production execution. Whether you are building agents with Anthropic's Claude Code, OpenAI's GPT-4, Google's Gemini, or local models using Ollama and LlamaIndex, standardized skill definitions ensure your agents behave predictably across different runtime environments.
What is the Model Context Protocol (MCP)?
The Model Context Protocol (MCP) is an open-source standard designed to connect LLMs to data sources, developer tools, and external environments. MCP establishes a bidirectional communication channel between client applications (like Cursor, Claude Desktop, or custom agent systems) and servers hosting data or capabilities. Standardizing instructions via SKILL.md enables LLMs to query databases, read local files, execute terminal commands, and integrate third-party APIs. SkillMD allows you to find ready-to-run MCP servers and prompt instructions for various occupations and technical tasks.
The Structure of a Professional SKILL.md File
A valid SKILL.md configuration is designed to be easily read by humans and parsed by LLMs. It contains precise system instructions, trigger conditions, required parameters, and execution examples. Below is the typical architectural blueprint of a professional agent skill:
- Metadata & Core Scope: Declares the name of the skill, author details, target models, and a description of the capability.
- Triggers & Intent Detection: Details semantic triggers that help the agent decide when to invoke this skill.
- System Prompts: Explicit system-level instructions that direct the agent's behavior, personality, safety guardrails, and formatting preferences.
- Capabilities & Tools: Lists the files, databases, or APIs the agent must access to complete the tasks.
- Few-Shot Examples: Demonstrates real inputs and outputs, helping the model generalize behavior through in-context learning.
Optimizing Agent Workflows for Modern LLMs
Writing effective agent skills requires deep knowledge of prompt engineering. With the release of advanced reasoning models like Claude 3.5 Sonnet, ChatGPT o1, and DeepSeek-V3, prompt templates must focus on structured thinking. Developers are encouraged to use XML tags (e.g., <thought>, <context>, and <rules>) to isolate execution boundaries. Standardized prompts prevent agents from suffering from context drift, ensuring that long-running tasks remain aligned with the initial system parameters.
Exploring by SOC Occupations and Creator Profiles
What makes SkillMD unique is its taxonomy. Instead of simple text search, we parse and organize files according to the Standard Occupational Classification (SOC) system. This means you can discover skills written for Computer and Mathematical roles, Business and Financial operations, Legal, Design, and and Educational Instruction fields. By tracking creator profiles, developers can study how different teams organize their custom instructions, compare version updates, and fork public configs for specialized enterprise use cases.
SkillMD operates as a high-performance index running on a fast Go backend and a highly responsive Astro SSR frontend. All search queries execute in milliseconds, featuring smart debouncing to prevent multiple API requests while keeping user data secure. Join our community of developers to standardize your AI agent instructions and optimize your LLM prompting workflows today.
Frequently Asked Questions
A practical guide to agent skills: what they are, how to inspect them, and how SkillMD helps you explore the ecosystem.