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...
foundry-hosted-agents
by aiappsgbbDeploy + manage Foundry hosted agents — GA June 2026: MAF 1.8.0, --deploy-mode code, agent.manifest.yaml, WS invocations, Foundry User + Project Manager roles. Read the body for SDK patterns, identity wiring, runtime selection, version rollout patterns, troubleshooting — do not deploy from this summary alone. USE FOR: deploy foundry agent, hosted agent, container agent, agent.yaml, agent manifest, code-mode deploy, FoundryChatClient, ResponsesHostServer, MAF, ACR push, batch eval, agent identity, Foundry User role, azd ai agent, entra-agent-id, WS invocations, Activity protocol, blue-green deploy, canary rollout, version rollback, traffic routing, version_selector, agent_endpoint. DO NOT USE FOR: prompt agents (use foundry-prompt-agents), ACA MCP (use foundry-mcp-aca), GHCP coding agent (use ghcp-hosted-agents), Citadel hub/spoke (use citadel-hub-deploy), pilot pipeline (use threadlight-deploy), continuous eval (use foundry-evals), Routines (use foundry-routines), A2A wiring (use foundry-toolbox).
gbb-humanizer
by aiappsgbbRemove signs of AI-generated writing from prose. Targets 29 patterns from Wikipedia's "Signs of AI writing" (em-dash overuse, rule-of-three, significance inflation, AI vocabulary, copula avoidance, false ranges, sycophantic openers, signposting, filler). Ships Microsoft GBB voice samples (seller pitch + technical blog) and density-preserving guardrails so domain lists and code blocks survive. Read the full skill body for the multi-pass procedure and pattern catalog. USE FOR: humanize prose, remove AI-isms, polish overview.html, polish demo script, polish speaker notes, AI tells, ChatGPT cadence, em dash overuse, rule of three, voice calibration, GBB seller voice, Cowork prose polish, threadlight overview polish, gbb-pptx speaker notes polish, sound less AI. DO NOT USE FOR: code, agent system prompts (directive style intentional), SKILL.md frontmatter, tables / KPI cards / code blocks, SME verbatim quotes, structured data.
gbb-pptx
by aiappsgbbGenerate professional PowerPoint (PPTX) presentations using python-pptx (the AI Apps GBB dark/light pitch-deck generator). USE FOR: create PowerPoint, generate PPTX, make slide deck, build presentation, convert markdown to slides, pitch deck, report as PPTX, create slides, gbb-pptx, gbb deck, dark-themed deck. DO NOT USE FOR: editing existing PPTX files (use the upstream `pptx` skill), PDF generation, Google Slides.
foundry-iq
by aiappsgbbBuild enterprise RAG into every threadlight process via Foundry IQ — Azure AI Search Knowledge Bases with agentic retrieval (multi-hop reasoning, query planning, citation-backed responses). DEFAULT knowledge retrieval pattern for every threadlight process; SPEC § 7 must declare a Knowledge Base for the process domain. USE FOR: knowledge base, RAG, agentic retrieval, policy assistant, citations, multi-hop QA, Knowledge Agent, AI Search Knowledge Base, document grounding, semantic retrieval, foundry-iq, knowledge index, hybrid search, vector search, kb-mcp, web iq, serverless knowledge base, purview acl knowledge. DO NOT USE FOR: structured-document extraction (use foundry-doc-vision-speech), MCP server deployment (use foundry-mcp-aca), agent runtime (use threadlight-deploy).
jwt-403-debug-expert
by aiappsgbbDiagnose 401/403 responses at the AI Citadel APIM gateway — decode JWT claims, verify Access Contract scope grants, and validate Foundry managed-identity token audience.
byok-401-debug-expert
by aiappsgbbDiagnose the silent BYOK 401 that Foundry hosted agents emit when "Foundry User" RBAC is assigned at PROJECT scope but missing at the underlying CognitiveServices ACCOUNT scope. Encodes the exact fix command.
citadel-hub-deploy
by aiappsgbbDeploy the **AI Citadel Governance Hub** (Layer 1) — APIM AI Gateway, Microsoft Foundry control plane, telemetry, 4 LLM APIs (Azure OpenAI, OpenAI Realtime, Universal LLM, Unified AI), private endpoints, access contracts. Wraps `Azure-Samples/ai-hub-gateway-solution-accelerator` branch `citadel-v1` (azd template) at a pinned commit. Ships 3 profiles (pilot-quickstart, enterprise-baseline, vnet-isolated-spoke-aware) plus tenant-isolated workflow. USE FOR: deploy citadel hub, citadel governance hub, apim ai gateway, ai-hub-gateway-solution-accelerator, citadel-v1, llm backend pool, unified ai api, universal llm api, openai realtime api, citadel access contract, multi-region foundry hub, BYO vnet hub, BYO log analytics, foundry network injection, managed redis semantic cache. DO NOT USE FOR: connecting a spoke to a hub (use citadel-spoke- onboarding), in-process governance (use foundry-agt), single-resource Foundry (use foundry-vnet-deploy or microsoft-foundry), tenant isolation (use azure-tenant-isolation).
aoai-model-migration
by aiappsgbbMigrate Azure OpenAI applications from GPT-4o/GPT-4o-mini to newer models (GPT-4.1, GPT-5, GPT-5.1 through GPT-5.4, o-series). Covers API changes, client configuration, parameter adaptation, prompt adjustments, and authentication. USE FOR: migrate model, switch model, upgrade model, GPT-4o replacement, AzureOpenAI to OpenAI client, v1 API, max_completion_tokens, reasoning_effort, developer role, system role, parameter adaptation, client factory, model classification. DO NOT USE FOR: retirement dates or lifecycle planning (use aoai-model-lifecycle), evaluation or A/B testing (use aoai-migration-evaluation).
gbb-pulse
by aiappsgbbDraft signals for the weekly AI GBB Pulse. Use when asked to write a pulse signal, customer win/loss, escalation, compete signal, product signal, IP initiative, or skills/people signal.
research-company
by aiappsgbbEmit a thin org-brief YAML profiling a target organisation — the company-specific overlay that pairs with an industry primer to drive a digital-clone-grade substrate fork. The brief captures only what the vertical can't infer: identity, ownership, size, geography, ~10 subsidiaries, the named ELT, 3–5 strategic themes, and any stack overrides the company publicly disclosed. Function tree, regulator set, entity kinds, proposed-domain library, rituals, KPIs all come from the matching industry primer. Four-phase procedure; claims carry confidence + sources[]; gaps go in uncertainties[], never invented. USE FOR: profile a company before a customer pilot, generate an org-brief, prepare a customer-flavoured demo fork. DO NOT USE FOR: single-process design (use threadlight-design), pitch-deck authoring, code generation, exhaustive personae harvest (compose-org expands archetypes via faker at fork time).
compose-org
by aiappsgbbFork an agentic substrate into a customer-flavoured digital clone using an org-brief YAML (from research-company) + the matching industry primer. Clones the substrate to a sibling repo, rebrands literal tokens, repacks the data-fabric generators against the brief's subsidiaries + named ELT, swaps the Kuzu entity-kind tables per the primer, extends the domain registry with the primer's workflow library, generates personae (ELT named from the brief, archetypes from the primer), seeds cadenced rituals + narrative arcs, and scaffolds Node MCP mocks for the brief's stack overrides. Local-only fork by default; no GitHub push. Refuses dirty trees, idempotent re-runnable. USE FOR: fork the substrate for a named customer (Telco, FSI, Airline, Retail, OEM, …), produce a digital-clone-grade demo repo. DO NOT USE FOR: incrementally adding one domain (use compose-domain inside the fork — planned, not yet available), authoring a new substrate from scratch, pitch decks.
threadlight-consumption-iq
by aiappsgbbUse after threadlight-safe-check --phase post-deploy returns green, before threadlight-production-ready, to project per-resource monthly Azure cost at the customer's declared production load and compare against 2–3 alternative SKUs per resource so the seller / SE can pick the cheapest config that still meets declared constraints before the customer signs off and turns on production traffic. Walks the deployed Bicep + `azd env`, reads SPEC § 12 `load_profile{}` (interactive wizard fills it on first run + writes back to SPEC), hits the public Azure Retail Prices API (`prices.azure.com`) directly via stdlib urllib with a versioned fixture fallback, and emits `docs/cost-projection.md` (human-readable scorecard with side-by-side SKU comparisons + mermaid cost share donut + top-N recommendations) plus `specs/cost-manifest.json` (strict v1 schema consumed by `threadlight-production-ready`'s cost pillar). Covers AOAI model deployments (PAYG vs PTU vs regional vs model swap), Foundry hosted-agent tiers, ACA SKUs (Cons
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.