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...
provider-uri-backfill
by mholziRecurring backfill job (#1289) that systematically fills missing per-provider track URIs across Beatify playlists. For every song that has a Spotify `uri` but is missing `uri_apple_music` / `uri_tidal` / `uri_deezer`, it resolves the gaps via the keyless Odesli / song.link API (Tidal + Deezer reliably, Apple Music best-effort) with a verifying Deezer-ISRC fallback; missing `uri_youtube_music` is filled via the YouTube Data API behind a resume-cursor + daily quota budget. Rate-limit safe (Odesli throttle + 429 backoff, YouTube daily budget). Emits a per-playlist Markdown coverage report. Use this skill whenever the user asks to backfill provider URIs, fill missing Apple/Tidal/ Deezer/YouTube links, improve provider coverage, run the URI backfill, or check per-playlist provider coverage.
dogfood-dtp
by mholziTechnical behavior test harness for Processminer's DTP Enhancer — the procedure-document (DTP) compare / regenerate / summary flow. Takes a process that already has a worked As-Is and an original ingested DTP, and drives it through dtp-regenerate, dtp-compare and dtp-summary via the running app's UI, exercising the finding-disposition paths (Open / Fix in DTP / Reconcile wiki), asserting that responses reflect correctly in the DTP Enhancer UI and the runtime store (`dtpReports`), and benchmarking speed per turn. Run ONLY when the user explicitly invokes /dogfood-dtp in the CLI. Never auto-route to it from the app's chat.
comment-review
by mholziWork through the open discussion comments on a wiki element with the SME — evaluate each comment's impact, incorporate the changes you agree on into the element, then post a closing summary into the discussion as the section's analyst. Use this whenever the user wants to review, act on or resolve the comments or notes left on an element.
run-lint
by mholziRun a lint pass over a process wiki — the consistency checkpoint. Check every element against its schema template, then sweep the whole process from all five perspectives for cross-section discrepancies and clarifying questions. Write the findings to the runtime store for the app's Review panel and re-open any approved element a finding implicates. Use this whenever the user asks to lint, run a lint pass, check the wiki for consistency, or review a process for issues.
dogfood-target
by mholziTechnical behavior test harness for Processminer's Target / Transformation track. Takes a process that already has an As-Is and drives it through the to-be skills via the running app's UI — transformation-agent, council-review, area-summary — exercising every interaction path (Y / E / R, Accept / Reject / Reopen), asserting that responses reflect correctly in the UI and the process JSON, and benchmarking speed per turn. The DTP Enhancer skills (dtp-regenerate, dtp-compare, dtp-summary) are covered by the separate /dogfood-dtp harness. Run ONLY when the user explicitly invokes /dogfood-target in the CLI. Never auto-route to it from the app's chat.
dtp-compare
by mholziCompare a chosen DTP (procedure document) against the corrected As-Is wiki and critically review it — surface every material discrepancy between the document and the analysis, and store the findings via the writeDtpComparison tool. This is review-only: no DTP is regenerated and no new artifact is written. Non- interactive: no SME questions, no approval loop. Invoked by the DTP Enhancer's "Select a source DTP" action. Use this whenever the user wants to compare, review or check an existing DTP against the wiki without rebuilding it.
transformation-agent
by mholziRun an interactive session with a banking subject-matter expert to develop the Target Process of a process — the target state / to-be design, the transformation decisions to reach it and the gaps to close — into the file-backed process wiki as draft elements. Use this whenever the user wants to design, refine or lock in the target/to-be state, plan the transformation, weigh transformation decisions or work the gap resolution of a process — even if they don't say "transformation agent".
dtp-regenerate
by mholziRegenerate a process's DTP (procedure document) from the corrected As-Is wiki and critically review the original ingested DTP against it. Read the original document and the worked As-Is, rewrite the procedure from the wiki's current truth, surface every material discrepancy between the old document and the analysis, and store both via the writeDtpReport tool — a new versioned .md artifact plus the critical-review findings. Non-interactive: no SME questions, no approval loop. Invoked by a button. Use this whenever the user wants to regenerate, rebuild or re-issue the DTP from the As-Is, or critically review the original DTP against the wiki.
control-compliance-specialist
by mholziRun an interactive elicitation session with a banking subject-matter expert to extract and document the risk & regulatory perspective of a process — controls, regulations, compliance gaps and audit findings — into the file-backed process wiki as draft elements. Use this whenever the user wants to document the controls of a process, capture regulatory obligations, map compliance gaps or audit findings, or run a control / risk / compliance extraction session — even if they don't say "control compliance specialist".
area-summary
by mholziGenerate an executive summary of one area of a process wiki — As-Is Process, Risk & Compliance, Client Experience, Innovation, Target Process or IT Architecture — written as an Amazon-style narrative memo. Read every section in the area and write the memo into the process's `summaries` field for the app to render. Non-interactive: no SME questions, no approval loop. Invoked by a button. Use this whenever the user wants an executive summary of an area.
foundational-run
by mholziWalk a freshly-ingested process end to end as a meticulous process analyst — read the whole wiki, then challenge every current-state element in foundational order to tease out rework, and surface the pain points and missing controls a source document never states, approving each with the SME. Resumable: a stopped run picks up where it left off. Use this after a document has been ingested into a process, or whenever the user asks to start, resume or run the foundational run / foundational review of a process.
source-regulation
by mholziAutonomously source the regulatory perspective of a process from the web — search for the financial-services regulation, supervisory rules and guidance that govern the process, then fill the Regulation section with draft elements. Non-interactive: no SME questions, no approval loop. Invoked by a button or another skill. Use this whenever the user wants to auto-source, web-source or pre-fill the regulations that apply to a process.
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.