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...
scitex
by ywatanabe1989[WHAT] Umbrella package that lazy-re-exports every standalone SciTeX sister package (io, plt, stats, scholar, writer, cloud, container, ...) under one `import scitex` namespace. [WHEN] Use when the user wants the all-in-one entry point, or asks for `import scitex`, `scitex.io`, `scitex.plt`, `@scitex.session`, or any `scitex.<sub>` access; defer to the relevant sister-package skill for sub-module specifics. [HOW] `pip install scitex` then `import scitex`; sub-modules load lazily on first attribute access (see `src/scitex/__init__.py`).
stx-cli
by ywatanabe1989Command-line interface for SciTeX platform operations — 30+ subcommands with lazy loading. Specialization of scitex-cli-convention.
crossref-local
by ywatanabe1989[WHAT] Offline, zero-API-key DOI lookup + full-text search over the CrossRef corpus. [WHEN] Use when the user asks to "look up a DOI", "resolve DOI to BibTeX", "find a paper by DOI", "enrich BibTeX with missing fields". [HOW] `import crossref_local` then call `habanero.Crossref()`.
scitex-writer
by ywatanabe1989[WHAT] End-to-end LaTeX manuscript toolchain — 45 MCP tools — project (clone/info/get_pdf), compile (manuscript / supplementary / revision with tracked changes), BibTeX (add/list/get/remove/merge, dedup by DOI), figures + tables (add/list/remove/archive, csv_to_latex, pdf_to_images), claims (`\vclaim{}` linked to scitex-clew session hashes for verifiable assertions), checks (float order, references), export + Overleaf migration, per-journal guidelines, AI2 Asta prompts. [WHEN] Use whenever the user asks to compile manuscript / build PDF / supplementary / revision with tracked changes, add figure/table/bibentry, merge or dedup . [HOW] `pip install scitex-writer` then `import scitex_writer`; see leaf skills for details.
scitex-scholar
by ywatanabe1989[WHAT] Scientific-paper search, metadata enrichment, PDF download, and BibTeX library management for the SciTeX ecosystem. [WHEN] Use when searching the literature, resolving DOIs, enriching citations, downloading PDFs through institutional access, or managing a reproducible paper library. [HOW] `import scitex_scholar` for the Python API; see leaf skills for entry points.
scitex-notification
by ywatanabe1989One-call alerting across 9 backends — audio TTS (spoken notification), desktop popup, emacs minibuffer, matplotlib banner, playwright browser toast, email (SMTP), webhook (HTTP POST), Telegram message, and Twilio phone call / SMS — with automatic fallback (default order: audio → emacs → matplotlib → playwright → email) and level-based routing (info / warning / error / critical can trigger different backend sets). Drop-in replacement for ad-hoc `smtplib.sendmail`, `requests.post(slack_webhook, ...)`, `plyer.notification.notify`, `twilio.rest.Client().calls.create`, `python-telegram-bot`, and hand-rolled "print + beep + email" patterns. Use whenever the user asks to "notify me when this finishes", "alert me if training fails", "send me an email when done", "call my phone if the server goes down", "text me the result", "push a Telegram message", "beep when the job completes", "escalate to phone call on critical errors", "ping Slack / webhook", or is wiring up notifications from scripts, pipelines, or AI agents.
scitex-ssh
by ywatanabe1989[WHAT] Persistent, auto-reconnecting SSH reverse tunnels for NAT traversal — installs an `autossh` systemd unit on the local host so a bastion/relay server can SSH back in even through firewalls and dynamic IPs. [WHEN] Use whenever the user asks to "set up a reverse SSH tunnel", "keep SSH alive through NAT", "access a lab machine from outside", "tunnel through a bastion", "autossh systemd service", "check tunnel status", "remove a tunnel", "expose this machine via a jump host", or mentions bastion server, NAT traversal, autossh, reverse SSH, HPC login node. [HOW] `pip install scitex-ssh` then `import scitex_ssh`; see leaf skills for details.
scitex-audio
by ywatanabe1989[WHAT] Unified text-to-speech (ElevenLabs / LuxTTS / gTTS / pyttsx3 with automatic fallback) AND local speech-to-text (whisper.cpp — tiny / base / small / medium / large-v3-turbo models). Smart local/relay routing so headless servers or WSL machines play audio on your laptop via a lightweight relay, SSH-tunnelable. Sequential playback queue prevents audio overlap when multiple agents speak concurrently… [WHEN] Use whenever the user asks to "say this", "speak", "play this text aloud", "voice notification", "read this out loud", "TTS this", "generate an audio file of …", "transcribe this audio", "speech to text", "whisper this . [HOW] wav / .mp3", "convert audio to text", "play audio on my laptop from the server", "check WSL audio", "announce the branch / directory", or needs voice I/O in any form.
scitex-hpc
by ywatanabe1989[WHAT] Generic SLURM dispatch + persistent reservations for the SciTeX ecosystem — one-shot `srun`/`sbatch`/`sync`/`poll`/`fetch_result` plus a `Reservation` primitive that books a node once and runs many short commands inside the allocation via `srun --jobid --overlap`, cutting queue wait from minutes to one ssh round-trip per command. [WHEN] Dispatching jobs to an HPC cluster from a laptop or login node — especially when iterating in tight loops, running multi-agent fleets, or doing jupyter-on-HPC where queue wait dominates wall time. [HOW] `from scitex_hpc import srun, sbatch, Reservation`, or `scitex-hpc <verb> ...`. Bastion-initiated SSH only; no persistent daemons.
scitex-os
by ywatanabe1989[WHAT] Host-check helpers (`check_host`, `is_host`, `verify_host`) and a path-aware safe `mv(src, dst)`. [WHEN] Use at the top of host-specific scripts instead of `socket.gethostname() == 'X'` checks; use `mv` for cross-filesystem renames. [HOW] `import scitex_os; scitex_os.check_host("bm198")` (raises) or `scitex_os.is_host("bm198")` (bool); `scitex_os.mv(src, dst)` (atomic same-fs, copy+unlink across devices).
scitex-introspect
by ywatanabe1989IPython-style introspection for any Python package. `q(obj)` mirrors IPython's `obj?` (signature + type hints + docstring summary). `qq(obj)` mirrors `obj??` (full source). `dir(obj)` lists attributes/methods with smart filtering. `list_api(module)` walks the module tree and returns the recursive public API as a structured tree. `get_docstring(obj)`, `get_exports(module)` (returns `__all__` contents), `find_examples(module)` (scans tests/ and examples/ for usage). Advanced: `get_class_hierarchy(cls)` (MRO + subclasses), `get_call_graph`, `get_dependencies`, `get_imports`, `get_signature`, `get_type_hints`. Drop-in replacement for opening a REPL, importing the target, and running `inspect.getsource(...)` by hand. Use whenever an agent needs to understand an unknown package's public surface, find usage examples without grepping, or inspect a class hierarchy without spelunking source files.
scitex-ui
by ywatanabe1989[WHAT] Shared frontend framework for SciTeX web apps — vanilla TS workspace shell + optional React components + Django static-asset integration. [WHEN] Use when user asks to build a SciTeX workspace app, add panel resizing, share a React DataTable, theme with design tokens, set up Django static asset discovery, inspect a DOM element, or mentions initShell, usePanelResize, Bridge infrastructure. [HOW] `pip install scitex-ui` then `import scitex_ui`; see leaf skills for details.
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.