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...
approvals-digest
by rockstarr-moonThis skill should be used when the scheduled daily run fires at 6 am local time, or when the user says "send the approvals digest", "what's waiting on me", "email me my approval queue", or "run the digest now". Scans /rockstarr-ai/03_drafts/ for files with approval_status: pending, sorts most-recent first by mtime, and emails one daily summary via send-notification. Email body follows skills/_shared/references/client-facing-output-voice.md — natural-language per-item cards with the channel as a plain-English noun and a one-line "what it is" line, no stop-slop scores or classification fields exposed. Exits silently when nothing is pending. Cross-bot by design — surfaces drafts from content, reply, outreach-*, and any future drafting bot without bot-specific code.
mark-booked
by rockstarr-moonThis skill should be used whenever a lead has booked a meeting — whether the bot booked it via book-meeting, or the client booked it manually by phone, email, or directly through their calendar. Trigger phrases: "mark this lead as booked", "Jane booked a meeting", "I booked this lead manually", "log the booking". This is the single source of truth for booking state: it flips Leads.state=booked, cancels every pending task for that lead, and writes a Replies row with classification=booked. Both the automated and manual booking paths converge here.
apply-label
by rockstarr-moonThis skill should be used after send-message returns success in the per-reply pipeline, when rockstarr-reply:present-for-approval's let-it-hang option routes to label-only, or when the user says "label this thread", "tag this lead as Not Interested", or "apply a label". Uses the Interceptly Labels UI via Chrome MCP (real coordinate clicks, not synthetic label.click()) to toggle the proposed label on the current thread, with Labels-nav-bug recovery if the UI navigates away to Campaigns. Refuses to apply more than one label per thread per pipeline run — labels are single-value.
intake-icp
by rockstarr-moonThis skill should be used when the user asks to "run the ICP interview", "capture the ideal client profile", "build the ICP", "run the perception gap exercise", or when run-intake dispatches to the ICP step. Walks the client through Phase A (7-step ICP interview) and Phase B (7-step Perception Gap exercise), one question at a time in the unified intake voice. Supports multiple ICPs via explicit loop. Checkpoints to /00_intake/intake/icp.md. Phase A output is the ICP section of client-profile.md; Phase B feeds the Positioning section.
invite-page-followers
by rockstarr-moonThis skill should be used when the scheduled monthly run fires (default 2pm second Tuesday), or when the user says "invite page followers" or "run the monthly page-follow invites". Drives the LinkedIn company-page admin dashboard via Chrome MCP, verifies the signed-in admin via profile-photo alt-text gate, reads credit balance, selects up to the configured invite target, clicks Invite, logs to /05_published/social/page-invites/YYYY-MM.md. Cycle-deduplicated — refuses to fire twice in the same credit cycle without force_rerun. Credits are separate from the daily-connect 20/day + 100/week cap.
ops-weekly-report
by rockstarr-moonThis skill should be used every Friday end-of-day, or when the user says "run the ops weekly report", "close out the ops week", or "weekly ops rollup". Aggregates the week's data from /05_published/ops/ + ops-mirror.xlsx into /06_reports/weekly/ops-[YYYY-WW].md. Sections: sales calls prepped, audits run (play breakdown + override rate), reengagements sent + replied, post-call processings (CRM-write success rate), deliverability trend with low-score callouts, and a stale-review-items section across ALL clients in one place.
backup-workbook-interceptly
by rockstarr-moonThis skill should be used every Friday end-of-day, or when the user says "back up the outreach mirror", "snapshot outreach-mirror.xlsx", or "save a weekly outreach backup". Copies the current outreach-mirror.xlsx to /06_reports/data/outreach-mirror-backup-YYYY-WW.xlsx, preserving the shared audit mirror (written by this plugin + rockstarr-reply) as it stood at week's end for rollback and comparison.
outreach-weekly-report-interceptly
by rockstarr-moonThis skill should be used after metrics-weekly-interceptly finishes, or when the user asks to "generate the weekly outreach report", "write this week's report", or "render the Interceptly weekly". Consumes Metrics (Weekly) rows for the target ISO week and produces a human-readable markdown report at /06_reports/weekly/outreach-YYYY-WW.md with per-account tables, week-over-week deltas, stale review-reply callout, non-ICP log highlights, flagged-leads list, session-failure summary, and a 'what Rachel / Jon should notice' block.
backup-workbook
by rockstarr-moonThis skill should be used every Friday end-of-day, or when the user says "back up the outreach workbook", "snapshot outreach-tasks.xlsx", or "save a weekly outreach backup". It copies the current outreach-tasks.xlsx to /06_reports/data/outreach-tasks-backup-YYYY-WW.xlsx, preserving the full workbook as it stood at week's end for rollback and audit.
outreach-weekly-report
by rockstarr-moonThis skill should be used after metrics-weekly finishes, or when the user asks to "generate the weekly outreach report", "write this week's report", or "render the Sales Nav campaign weekly". It consumes Metrics (Weekly) rows for the target ISO week and produces a human-readable markdown report at /06_reports/weekly/outreach-YYYY-WW.md with per-campaign tables, week-over-week deltas, stale review-reply callout, weekly cap usage, and a "what Rachel / Jon should notice" block.
fill-week
by rockstarr-moonThis skill should be used when the user asks to "fill the week", "build next week's social batch", or when the scheduled Friday-morning run fires. Builds a balanced weekly batch of 5-10 short-form posts: reads social_posts_per_week + social_mix from stack.md, picks topics from first-party KB + recently approved long-form, dedups against the publish log, loops draft-social per slot. Writes a batch manifest week-YYYY-WW.md in 03_drafts/social/ plus per-post drafts. Batch is the approval unit; once approved, run publer-export.
publer-export
by rockstarr-moonThis skill should be used when the user asks to "export to Publer", "build the Publer CSV", or "generate the scheduler import" for an approved batch. Reads the approved weekly batch manifest + per-post files and writes a Publer-shaped CSV (Date, Time, Account, Post, Media, Link, First Comment, Labels) to 05_published-staging/social/YYYY-WW.csv. Manual trigger only — does not auto-run on approval. Refuses if stack.md.social_scheduler is not 'publer'. Does not post — the client imports the CSV into Publer themselves.
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.