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...
npc-creator
by camaugerCreate complete TTRPG NPCs with narrative depth and mechanical stats for D&D 5e. Generates appearance, personality, desires, secrets, backgrounds, stat blocks (tiered by importance), portrayal cues (voice, mannerisms, GM tips), and plot hooks. Handles quick background NPCs, significant social characters, major combat-capable characters, and everything between. Use when GM needs fleshed-out NPCs for campaigns, modules, or improvisation.
fanzine-advice
by camaugerCreate engaging fanzine-style advice articles (1000-2000 words) for tabletop RPG Game Masters and players. Use when the user requests DMing advice, GMing techniques, player guidance, table management tips, or practical RPG advice in the voice and style of classic zine culture - personal, opinionated, tested-at-the-table wisdom with a DIY aesthetic.
backgrounds
by camaugerCreate complete backgrounds for Dungeons & Dragons 5th Edition (2024 rules), formatted to match the 2024 Player's Handbook style. This skill produces mechanically balanced backgrounds with full narrative context, supplementary content, and compatibility checks against official materials.
ttrpg-cliche-buster
by camaugerAudite du contenu de supplément JDR (lieux, PNJ, factions, accroches d'aventure, monstres, rencontres) pour identifier les clichés, lieux communs et tropes usés, puis génère des alternatives créatives et des questions provocatrices pour forcer la sortie des sentiers battus. Bilingue français/anglais selon le contexte. Utiliser quand un créateur JDR veut améliorer l'originalité de son contenu, quand un texte "semble générique", quand l'utilisateur dit "qu'est-ce qui est cliché ici ?", "aide-moi à sortir du cliché", "rends ça plus original", "brise ce trope", "trouve les lieux communs dans mon texte", "pose-moi des questions pour améliorer ça", "casse cette idée", "subvertis ce concept", ou quand on crée des PNJ / factions / lieux / aventures / monstres qui semblent familiers. Déclencher aussi proactivement quand du contenu JDR soumis contient des signaux de cliché évidents (taverne de départ, vieux mentor sage, ancien mal qui se réveille, etc.).
magic-item-creator
by camaugerCreate complete custom magic items for D&D 5e 2024, formatted to match the 2024 DMG style. Covers the eight standard categories (Armor, Weapons, Potions, Rings, Rods, Scrolls, Staves, Wands, Wondrous Items) at all five rarities (Common, Uncommon, Rare, Very Rare, Legendary). Each item gets a stat block with category / rarity / attunement / properties / charges plus lore and origin paragraph, 1–2 variants, and 2–3 plot hooks. Use on "crée un objet magique", "create a magic item", "épée magique custom", "magic weapon for level 5", "wand of X", "homebrew ring", "Uncommon potion", "Rare wondrous item", "custom legendary item", "fabrique-moi un objet magique", "potion custom", "objet merveilleux", "anneau magique homebrew", "+1 sword with a twist", "magic item with a story". Boundary: cursed flavor and minor sentience can be added as options to standard items, but full Sentient items, full Cursed items, and Artifacts (unique campaign-tier) are deferred to dedicated future skills. NOT for non-magical equipment, pot
adventure-converter
by camaugerConvert TTRPG adventures, scenarios, supplements, and stat blocks between fantasy systems. Four paths: (1) D&D 5e 2014 → 5e 2024, (2) D&D 5e → Pathfinder 2e, (3) D&D 5e → OSR (B/X, OSE, Shadowdark, generic old-school), (4) D&D 5e → narrative-light (PbtA/FitD adaptation notes). For each: audit source, build conversion matrix, convert stat blocks and DCs, adjust encounter design, document kept/cut/added, deliver a publishable target-system document with GM notes. Use on "convert to 5e 2024", "port my 5e 2014 module to 2024", "5e to PF2e conversion", "OSR version of my scenario", "Shadowdark port", "convertis en 5e 2024", "porte en PF2e", "version OSR", "adapte en old-school". Boundary: converts an EXISTING adventure (use scenario-writer to create new content). NOT for cross-genre conversion (fantasy → sci-fi), NOT for systems outside the four supported paths (Call of Cthulhu, Vampire, Mörk Borg — bespoke handling), NOT for reverse conversion (target → source is a different discipline once warranted).
tarot-card-portrait
by camaugerGénère des prompts Midjourney complets pour illustrer un personnage TTRPG (ou original) sous forme de carte de tarot verticale et ornementale. Utiliser quand l'utilisateur veut créer une carte de tarot pour un personnage, illustrer un PNJ ou personnage joueur sous forme de tarot, assigner un arcane majeur à un personnage ou concevoir une illustration de tarot érotique/dark fantasy pour la marque Fantasy Vixens. Déclenche aussi sur : "fais-moi une carte de tarot pour [personnage]", "quel arcane représente [PNJ]", "génère le prompt Midjourney pour la carte de [personnage]", "illustre [personnage] en style tarot".
encounter-builder
by camaugerBuild single encounters for D&D 5e 2024 across all five types: Combat, Social, Exploration, Chase, and Downtime. For each: calibrate difficulty against the specific party (size, level, composition), design terrain and objectives including alternative win/loss conditions, and produce table-ready deliverables (read-aloud for entrance and resolution + GM cheat sheet with pre-calculated initiative or key DCs). Use on "build an encounter", "balance this combat", "design a social encounter", "create a chase", "downtime scene", "encounter for 4 PCs level 5", "calibrate XP budget", "fabrique-moi une rencontre", "design un combat pour mon groupe", "rencontre sociale avec enjeux", "poursuite", "scène de tension non-combat", "encounter with objectives beyond TPK", "alternative win conditions", "encounter terrain with hazards". Boundary: this skill produces ONE encounter at a time. For scenario-length chains of linked scenes (three-clue rule, node-based structure) use scenario-writer. For monster stat blocks and per-mons
monster-creator
by camaugerCreate complete monsters for D&D 5e 2024, Monster Manual format. Produces a full stat block (type, size, alignment, AC, HP, speeds, abilities, saves, skills, resistances, senses, languages, CR, traits, multiattack, actions, bonus actions, reactions) plus lair actions + regional effects + variants for CR 5+, legendary and mythic actions for CR 11+, ecology and habitat, signs of presence, GM tactics, and 3–5 plot hooks per creature. Single- monster mode, publication-ready. Use on "crée un monstre", "create a monster", "stat block 5e 2024", "custom monster", "design a creature", "monstre custom", "bestiaire D&D", "fabrique-moi un monstre", "stat block CR 7", "monster manual style", "boss for my campaign", "mini-boss level 5", "homebrew creature", "balance this monster", "creature with lair actions". Boundary with npc-creator: humanoid NPCs with personality and secrets stay in npc-creator; non-humanoid creatures or species treated as ecological entities (a hag coven as a species, a goblin warband as a stat block)
spell-creator
by camaugerCreate complete custom spells for D&D 5e 2024, formatted to match the 2024 Player's Handbook style. Covers all 10 spell levels (Cantrip → 9th), the eight schools (Abjuration, Conjuration, Divination, Enchantment, Evocation, Illusion, Necromancy, Transmutation), every standard class spell list, and all casting frames (Action, Bonus Action, Reaction, ritual, minutes / hours / 24 hours). Each spell ships with PHB-format stat block (level, school, casting time, range, components, duration, classes), description text, scaling at higher levels (or cantrip scaling), and 1–2 plot hooks tied to origin or notable users. Use on "crée un sort", "create a spell", "cantrip homebrew", "3rd-level evocation", "wizard spell for X", "concentration spell that does Y", "ritual spell", "fabrique-moi un sort", "sort original", "homebrew spell with twist", "spell for a Storm Sorcerer", "necromancy at 5th level", "BBEG signature spell". Boundary: this skill produces ONE spell per call (or a small thematic set of 2–3 related spells).
stat-block-validator
by camaugerAudit and validate D&D 5e 2024 monster stat blocks for mechanical correctness, format conformity, and internal consistency. Five sequential checks: (1) format compliance against MM 2024 layout, (2) arithmetic validation (modifiers, saves, attacks, AC, HP), (3) CR math audit (offensive DPR × 3 rounds + defensive effective HP + AC), (4) action economy audit (multiattack, recharge, legendary/lair/regional/mythic actions), (5) trait vocabulary audit (2024 phrasing, official conditions, no 2014 syntax). Produces a structured defect report with severity (Critical / High / Medium / Low) and corrected stat block. Use on "validate this stat block", "audit my monster", "check the CR on this", "is this stat block legal 2024", "verify monster math", "audit homebrew creature", "valide ce bloc de stats", "audite ce monstre", "vérifie le CR". Boundary: VALIDATES existing stat blocks (use monster-creator to design new ones). Targets D&D 5e 2024 exclusively (not 2014, not PF2e, not OSR — those need different audit patterns on
graph-api-quality-auditor
by camaugerAudit et améliore la qualité du code Python qui appelle Microsoft Graph API dans un contexte Azure. Utiliser cette skill quand l'utilisateur soumet du code Python faisant appel à Graph API (Microsoft 365, Azure AD, Teams, SharePoint, OneDrive, Bookings, etc.) et veut l'analyser, le corriger, ou l'améliorer. Déclencher aussi sur : "regarde mon code Graph API", "améliore mon script MSAL", "mon code Graph fait des erreurs 429", "comment paginer correctement avec Graph", "est-ce que mon auth est sécurisée", "mon token expire", "optimise mes appels Graph", "audit de mon code Microsoft Graph", "regarde mon automatisation Azure".
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.