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...
mtg-std-upgrade
by guuseImprove an existing 60-card Standard deck for MTG Arena (MTGA) that the user pastes in. Use this skill whenever the user already has a Standard/Arena decklist and wants to upgrade, tune, optimize, or tech it against the current ladder meta — rather than build a new deck from scratch. Triggers on phrases like "upgrade my Standard deck", "improve this Arena deck", "here's my list, what should I swap", "tune my mono-red for the meta", "what cards should I craft to improve this", or any request that pastes or references an existing 60-card Arena list and asks for improvements. The user pastes their current list inline. The skill diagnoses the deck (curve, mana base, consistency, meta matchups), then recommends the highest-impact swaps — built from cards the user already owns first (via their Arena collection export), and costed in Arena wildcards against a budget tier that, because this is an upgrade, is usually much lower than building from scratch. Verifies legality and rarity via Scryfall; reads the live meta
mtg-sync
by guuseKeep the user's MTG decks and collection in a private git repo so the same data follows them across machines and mobile. Use this skill to set up syncing the first time (ask for the user's private "mtg-data" repo and clone it), to pull the latest decks/collection before building or upgrading a deck, and to push newly built decks afterwards. The deck skills (mtg-edh-build, mtg-edh-upgrade, mtg-std-build, mtg-std-upgrade) invoke this automatically before and after they run — pull before, push after — the same way they rely on the mtg-db skill for card data. Syncing is best-effort: if git, a repo, or the network isn't available it reports that and the build proceeds with the local workspace. Triggers on "sync my decks", "store my decks somewhere", "set up my mtg-data repo", "use my decks on my other PC / phone", or any cross-machine deck/collection persistence request.
mtg-db
by guuseBuild or refresh the local Scryfall card database (.mtg/database/cards.sqlite) that the MTG deckbuilding skills read from instead of calling the Scryfall API. Use this skill when the user wants to set up, build, refresh, or update their local MTG card data, when card prices or sets seem out of date, or as the one-time setup step before building or upgrading Commander/EDH or MTG Arena Standard decks. The deck skills also invoke this automatically when the database is missing or stale. It downloads Scryfall's "Default Cards" bulk file, collapses it to one row per unique card (cheapest EUR/USD price, Arena availability, rarity, legalities, Game Changer flag, EDHREC rank), and stores it as SQLite for fast, mostly-offline querying — sharply cutting Scryfall API calls and rate-limiting.
mtg-edh-analyze
by guuseAnalyze and star-rate an existing 100-card Magic the Gathering Commander (EDH) deck that the user pastes in, judged against a target power bracket. Use this skill whenever the user wants their Commander deck rated, scored, graded, reviewed, or assessed — "how good is my deck", "rate my EDH deck", "give my deck a score out of 5", "how does this deck stack up at Bracket 3", "analyze my commander deck", "is this deck any good", "what bracket is my deck", or any request to evaluate a Commander list rather than build or upgrade one. The user pastes their current decklist inline and names (or is asked for) a target bracket (1–5). The skill measures the deck's hard stats (land count, mana curve, ramp/draw/interaction density, EDHREC-rank staple signal, Game Changer count, color-identity legality, price) via the local Scryfall database, reads every card's Oracle text to score synergy density, then awards an overall ★ rating (with a per-dimension scorecard) for how well the deck performs *at its bracket* — covering pe
mtg-edh-build
by guuseBuild a complete, balanced 100-card Magic the Gathering Commander (EDH) deck around a chosen commander. Use this skill whenever the user wants to build, brew, design, or upgrade a Commander/EDH deck, names a legendary creature and asks "build a deck around X", asks for a decklist for a commander, wants help picking cards that synergize with a commander, or wants a Commander deck within a budget or at a specific power bracket. Triggers on phrases like "build me a commander deck", "EDH deck for [commander]", "brew around [legendary creature]", "100-card singleton deck", "help me build my Atraxa deck", or any request that pairs a commander name with deckbuilding. The skill pulls proven cards from EDHREC and mtgdecks.net, fills gaps and prices everything via Scryfall (which carries Cardmarket EUR prices), and applies a disciplined 7-step methodology to produce a tuned list with per-card pricing and a budget/bracket target.
mtg-edh-upgrade
by guuseImprove an existing 100-card Magic the Gathering Commander (EDH) deck that the user pastes in. Use this skill whenever the user already has a Commander/EDH decklist and wants to upgrade, tune, optimize, fix, or "make it better" — rather than build a new deck from scratch. Triggers on phrases like "upgrade my commander deck", "improve this EDH deck", "here's my decklist, what should I change", "tune my Atraxa list", "what are the best upgrades for my deck", "help me cut/add cards", or any request that pastes or references an existing 100-card list and asks for improvements. The user pastes their current list inline. The skill diagnoses the deck against a proven 7-step methodology (land count, card advantage, ramp, interaction, curve, win conditions), then recommends the highest-impact swaps within a budget — and because this is an upgrade, the budget is usually much smaller than building a deck from scratch. Prices everything via Scryfall (Cardmarket EUR) and respects the deck's color identity and power bracke
mtg-std-build
by guuseBuild a 60-card Standard-format deck for MTG Arena (MTGA), centerpiece-first, tuned against the current ladder meta, and costed in Arena wildcards by rarity. Use whenever the user wants to build, brew, or netdeck a Standard deck for Arena, names a card and asks to build a Standard deck around it, wants a deck for the BO1 ladder or BO3 with a sideboard, asks what to craft for a budget number of wildcards, or wants a deck that beats the current Standard meta. Triggers on phrases like "build me a Standard deck", "MTGA deck for [card]", "brew around [card] in Standard", "budget Arena Standard deck", "what should I craft", or any request pairing Standard/Arena with deckbuilding. Every card is labeled with its rarity and the build respects a wildcard budget tier (1-5). Verifies legality and rarity via Scryfall, reads the live meta from untapped.gg/mtggoldfish, and outputs an Arena import list plus an annotated wildcard-cost breakdown. Starts by asking for the user's MTG Arena collection export and, when given one,
mtg-card-finder
by guuseCollaboratively find the right Magic the Gathering cards for a player by brainstorming with them and researching card text deeply. Use this skill whenever the user wants help finding cards rather than building a whole deck — when they want to pick a commander, fill a specific gap in a deck, add a category (more card advantage, ramp, removal, a finisher, protection), find synergy pieces for a theme or combo, or solve a problem their deck has ("my deck can't close games", "I keep running out of cards", "I die to board wipes", "what's the best removal in these colors"). Triggers on phrases like "help me find cards for…", "what commander should I play", "I need more card draw in my deck", "find me cards that synergize with X", "what should I add to fix Y", "best cards for this strategy", "recommend cards that do Z", or any open-ended card-discovery or deck-problem-solving request that is NOT a full build/upgrade. The skill starts by pinning down the *purpose*, gathers tailored context by brainstorming, then resea
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.