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...
search-external-sites
by PioneerAIAcademyGenerates search URLs for external genealogy sites (Ancestry, MyHeritage, FindMyPast, FindAGrave, Newspapers.com) and walks the user through the click-capture-analyze workflow. Logs each search to research.json (including nil results) and triages results from captured PDFs before passing records to record-extraction. GPS Step 1 — Reasonably Exhaustive Research (external site execution). Use when the user says "search Ancestry", "search MyHeritage", "search FindMyPast", "search FindAGrave", "search Newspapers.com", when a plan item targets a non-FamilySearch repository, or when the user uploads a PDF capture from an external genealogy site. Do NOT use when the target is FamilySearch (use search-records), when the user wants to plan what to search (use research-plan), or when the user wants to analyze a single record already in context (use record-extraction).
search-full-text
by PioneerAIAcademyExecutes full-text searches against FamilySearch AI-transcribed historical document images per the research plan. Uses the fulltext_search MCP tool with Lucene-style operators (+/-/"…"/?/*). Uniquely surfaces witnesses, neighbors, heirs, sureties, appraisers, and other non-principal mentions that indexed search misses. Logs every search including nil results and passes promising records to record-extraction. GPS Step 1 — Reasonably Exhaustive Research (full-text execution). Use when the user says "full-text search", "search for witnesses mentioning [person]", "search newspapers for [person]", "find [person] in deeds/probate/court minutes", when a plan item targets FamilySearch full-text search, when looking for FAN club (Family/Associates/Neighbors) mentions, when searching pre-1850 US records with thin indexed coverage, or when searching Latin American notarial records. Do NOT use when the target is a structured indexed search by person attributes (use search-records), when the target is Ancestry, MyHeritage
search-records
by PioneerAIAcademyExecutes searches against FamilySearch historical records per the research plan. Routes to the correct MCP search tool based on record type, triages results using match scoring, logs every search including nil results, and passes promising records to record-extraction. GPS Step 1 — Reasonably Exhaustive Research (execution phase). Use when the user says "search for [person]", "find [person] in [record type]", "execute the plan", "run the next search", "search FamilySearch", or when a plan item targets a FamilySearch repository. Do NOT use when the target is Ancestry, MyHeritage, FindMyPast, FindAGrave, or Newspapers.com (use search-external-sites), when the user wants to plan what to search (use research-plan), or when the user wants to analyze a record already found (use record-extraction).
search-wiki
by PioneerAIAcademySearch the FamilySearch Research Wiki for genealogy research guidance and save the findings as a markdown file in the user's working folder. Use when the user asks to "search the FamilySearch wiki", "check the FS research wiki", or asks a how-to research question — how to find a record type (birth, marriage, death, census, immigration, military, church, land, probate), how to research ancestors from a specific country or region, or how to use a FamilySearch resource or repository. Do NOT use when the user explicitly names Wikipedia (use search-wikipedia), wants a comprehensive locality records-availability guide (use locality-guide), or wants narrative historical background such as migration patterns or boundary changes (use historical-context).
search-wikipedia
by PioneerAIAcademyLook up a topic on Wikipedia — the general-purpose encyclopedia — and save the summary as a markdown file in the user's working folder. Use when the user explicitly names Wikipedia, or asks to look up or save general background on a topic, person, place, or historical event. Do NOT use when the user names the FamilySearch Research Wiki or "FamilySearch wiki" (use search-familysearch-wiki), wants a locality records-availability guide — what records exist for a place and where they are held (use locality-guide), or wants narrative genealogical history such as migration patterns or boundary changes (use historical-context).
tree-edit
by PioneerAIAcademyHandles direct edits to tree.gedcomx.json — adding facts, correcting values, creating persons, adding relationships, merging two persons confirmed to be the same individual, verifying that the tree already reflects a known fact (no-op confirmation), checking what records FamilySearch has matched or attached to a tree person, and checking whether a tree person has possible duplicates that may need merging. Use when the user says "correct this name", "change birth year", "add occupation", "merge these two persons", "fix this fact", "add a relationship", "this person's name is wrong", "verify the tree reflects this", "check the tree", "make sure the tree shows", "confirm this fact is in the tree", "what records are attached to this person", "what hints does FamilySearch have for this person", "check record matches for", "find possible duplicates for", "are there duplicate persons for", "check for merge candidates", when proof-conclusion requests a person merge after confirming identity, or when the user needs to
proof-conclusion
by PioneerAIAcademyWrites GPS-conformant proof conclusions — selects the confidence tier (Proved/Probable/Possible/Not Proved/Disproved), chooses the proof vehicle (Statement/Summary/Argument), and produces a self-contained narrative markdown that can be uploaded to FamilySearch. Updates tree.gedcomx.json when the tier reaches probable or higher. GPS Step 5 — Soundly Reasoned, Coherently Written Conclusion. Use when the user says "write the conclusion", "what's the proof?", "summarize the evidence", "write a proof statement", "write a proof argument", "conclude this question", when assertions and person_evidence exist for a question, or when a hypothesis reaches supported status. Do NOT use when the user wants to resolve a conflict (use conflict-resolution), wants to select the next question (use question-selection), or wants to classify evidence (use assertion-classification).
init-project
by PioneerAIAcademyInitializes a new genealogy research project with GPS-conformant file structures. Creates research.json (GPS audit trail) and tree.gedcomx.json (simplified GedcomX deliverable) from a FamilySearch person ID. If the user does not have a FamilySearch ID, searches the Family Tree by name using person_search to find the right person. Implements Steps 1-2 of the genealogical research process (define the problem and survey known information). Use when the user says "new project", "start research", "research [person]", "find parents of", "begin researching", "I don't have their FamilySearch ID", or provides a FamilySearch person ID to start working with. Do NOT use when a research.json file already exists in the folder — use project-status instead to resume an existing project.
interpret-e2e-result
by PioneerAIAcademyReads an e2e benchmark run log and explains what happened — verdict, stop reason, which expected findings the agent did and didn't recover, and the most likely failure cause (agent reasoning regression, /research routing regression, sub-skill regression, FamilySearch data drift, or single-run jitter). Points the user at the right transcript section to read next. Use when the user says "what happened in this e2e run", "interpret this e2e result", "why did this fixture fail", or "read the latest e2e runlog". Do NOT use to author or modify a fixture (use author-e2e-fixture), to interpret a unit-test scratch run (those are developer-facing — read the run log directly), or to grade a single research question in a live project (use the relevant analysis skills like timeline or conflict-resolution).
search-familysearch-wiki
by PioneerAIAcademySearch the FamilySearch Research Wiki for genealogy research guidance and save the findings as a markdown file in the user's working folder. Use when the user asks to "search the FamilySearch wiki", "check the FS research wiki", or asks a how-to research question — how to find a record type (birth, marriage, death, census, immigration, military, church, land, probate), how to research ancestors from a specific country or region, or how to use a FamilySearch resource or repository. Do NOT use when the user explicitly names Wikipedia (use search-wikipedia), wants a comprehensive locality records-availability guide (use locality-guide), or wants narrative historical background such as migration patterns or boundary changes (use historical-context).
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.