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...
mr-robot
by TykoDevPressure-tests the scoped surface from an adversarial viewpoint by tracing exploit paths, abuse cases, and chaining conditions. Use when the user asks to run adversarial review, pressure-test this code, think like an attacker, or look for exploit chains — even when they only ask "how would someone break this?". Focuses on chained, attacker-driven scenarios; hands single defensive flaws with no abuse sequencing to `review/security-review` and security-governance decisions to `review/cso`.
admiral
by TykoDevSupremeTeam primary entry orchestrator for design, build, review, ship, investigate, checkpoint/resume, and skill or team creation. Routes every delivery-lifecycle request under one intake, one save-protocol run, and one gatekeeper. Use when the user says: run the full pipeline, ship end to end, resume from a checkpoint, design/build this project, review or audit this codebase, find the root cause, create a skill, build a team, or run Admiral. All lifecycle work enters here first; standalone utility tools run directly.
gatekeeper-admiral
by TykoDevAdmiral-pipeline cross-stage gatekeeper, normally invoked by `admiral` at each boundary; if reached directly without an active Admiral handoff, hand off to `admiral` first (see routing-doctrine.md). Validates cross-stage packages and decides whether they are ready to advance between major delivery boundaries. Use when `admiral` routes a boundary package for a verdict, or the user asks to validate the handoff, check build readiness, review delivery readiness, challenge the package boundary, or verify whether a package can move from one orchestrator to the next.
gatekeeper-build
by TykoDevValidates build-phase deliverables, hardening evidence, and completeness claims before code work can advance. Use when the user asks to validate the build deliverable, review build phase output, check build readiness, challenge this build packet, or verify whether implementation evidence is ready for review — even when they only ask "is the build ready?". Gates the build→review boundary only; defers the other gates to `design/gatekeeper-design`, `review/gatekeeper-code`, and `gatekeeper-admiral`.
gatekeeper-code
by TykoDevValidates consolidated review packages and decides whether the review evidence is ready for delivery or needs another round. Use when the user asks to validate the review package, check review readiness, challenge this review packet, or gate the review output — even when they only ask "is this review done?". Gates the review→delivery boundary only; defers the other gates to `build/gatekeeper-build`, `design/gatekeeper-design`, and `gatekeeper-admiral`.
gatekeeper-design
by TykoDevValidates design-phase deliverables before a design package can advance to the next design activity or leave the design pipeline. Use when the user asks to validate the design deliverable, review design phase output, check design readiness, challenge this design packet, or verify whether the design package is coherent enough for build consumption — even when they only ask "is the design done?". Gates the design→build boundary specifically; defers the build→review gate to `build/gatekeeper-build`, the review→delivery gate to `review/gatekeeper-code`, and the cross-stage delivery gate to `gatekeeper-admiral`.
commander
by TykoDevAdmiral-pipeline design sub-orchestrator, normally invoked by `admiral`; if reached directly for lifecycle work without an active Admiral handoff, hand off to `admiral` first (see routing-doctrine.md). Runs the design pipeline from initial scope to an approved design package with requirements, plans, architecture, interface contracts, and implementation guidance. Use when `admiral` delegates the design boundary, or the user asks to design this system, create the design package, start the design pipeline, or plan and architect this project.
investigate
by TykoDevAdmiral-pipeline investigation component, normally invoked by `admiral`; if reached directly for lifecycle work without an active Admiral handoff, hand off to `admiral` first (see routing-doctrine.md). Runs disciplined root-cause analysis across code, logs, runtime clues, and environmental evidence when the failure shape is still unclear. Use when `admiral` delegates an investigation, or the user asks to investigate this issue, find the root cause, explain why this broke, trace the failure, or narrow a messy incident down to one credible explanation.
session-memory
by TykoDevMemory component of the Admiral delivery pipeline: captures checkpoints and reusable learnings so long-running delivery work can resume safely and later phases can query prior discoveries. Use when `admiral` engages it for a checkpoint, or the user asks to save progress, checkpoint this run, resume from saved state, or record a learning — even when they only say "save where we are" or "remember this for later".
skill-maker
by TykoDevEnd-to-end orchestrator for creating, reviewing, improving, optimizing, and packaging Claude skills and coordinated skill teams. Use when the user says "create a skill", "make a skill", "build me a skill", "write a skill", "run the skill pipeline", "review this skill", "harden this skill", "take this skill to 100", "ship this skill", "make it production-ready", "create and review a skill", "iterate this skill to perfection", or describes a desired skill behavior without naming skill-maker. Also use when `admiral` delegates skill or team creation. Routes all drafting, evals, fixes, rubric scoring, and packaging to specialists; do not use for general code review, architecture, or non-skill authoring tasks.
gatekeeper-azure
by TykoDevThis skill should be used when the user asks to "validate Azure deliverables", "gate-check the Azure pipeline", "challenge the Bicep design", "validate deployment results", "approve this Azure phase", "quality-gate the Azure output", "is this Bicep ready?", "pressure-test this Azure plan", "double-check this deployment evidence", or "review the Azure configuration". Single adversarial gate validating deliverables from all Azure pipeline specialists. Produces no infrastructure artifacts or deployments — it only challenges, validates, approves, revises, or escalates Azure deliverables using the canonical pipeline verdicts `APPROVED`, `REVISE`, and `ESCALATE`. DO NOT USE for designing infrastructure (use azure-architect). DO NOT USE for deployment execution (use azure-deployer). DO NOT USE for cross-pipeline gating (use gatekeeper-admiral).
code-review
by TykoDevThis skill should be used when the user asks to "review this code", "do a code review", "review this pull request", "check this PR", "evaluate code changes", "review for design and complexity", "assess code readiness for merge", "review this diff", "give feedback on this code", "look at my changes", "what would you comment on here?", "is this code okay?", or "is this ready to merge?". Performs comprehensive code review covering design, functionality, complexity, tests, naming, comments, style, and documentation using Google's 8-dimension framework with risk-tiered PR assessment (Low/Medium/High/Critical). DO NOT USE for bug hunting (use bug-review). DO NOT USE for security-specific review (use security-review). DO NOT USE for frontend visual audit (use design-qa or frontier).
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.