Skills

Symphony by OpenAI turns project work into isolated, autonomous agent implementation runs that monitor task boards, spawn coding agents, and land PRs with proof of work, letting teams manage work instead of supervising agents.

from lqshow/linyuebanzi-skills

A PostHog skill that guides AI agents through setting up LLM analytics, part of PostHog's official skills collection covering analytics integration, feature flags, error tracking, and LLM observability, auto-generated by context-mill.

A drop-in Claude Code configuration layer providing 119 domain skills, 59 specialist agents, 30 slash commands, 10 lifecycle hooks, and 12 MCP server configs that auto-activate based on your project context.

from lilmgenius/paperthin

A collection of 148+ ready-to-use scientific and research skills for AI agents, now including financial/SEC research, U.S. Treasury fiscal data, and market data tools alongside biology, chemistry, medicine, and machine learning workflows. Compatible with Cursor, Claude Code, and Codex via the Agent Skills standard.

MCP server development guide skill for creating high-quality Model Context Protocol servers in Python with FastMCP or Node/TypeScript with MCP SDK for LLM integration with external services

Part of ClawBee, a family meal planner skill set for OpenClaw that lets you photograph your fridge, generate a 3-day meal plan, and receive a shopping list on Telegram, with integrations for Discord, price searching, and an SQLite-backed pantry tracker.

Part of a comprehensive 99-skill library covering the full web lifecycle, this skill handles media asset management within a broader suite for brand building, design, content, SEO, development, and growth.

AI research engineering skill providing guidance on MILES reinforcement learning training, from a comprehensive open-source library of research skills for AI agents.

A Nimble web data toolkit skill that provides web search, data extraction, site mapping, and crawling capabilities via the Nimble CLI for Claude Code and other AI agents.

from ascend/agent-skills

Part of OpenAkita, an open-source multi-agent AI assistant where multiple specialized agents collaborate in parallel. Features 89+ tools, 30+ LLM providers, 6 IM platform integrations, plan mode, ReAct reasoning, and a GUI-based 5-minute setup.

An open-source multi-agent AI assistant with organization orchestration, plugin system, sandbox security, 30+ LLM support, 6 IM platform integration (Slack, Discord, WeChat, etc.), and 89+ built-in tools.

Over 3,000 app-specific integration skills by Membrane that teach AI agents to connect with Gmail, Slack, HubSpot, Salesforce, and thousands of other apps. Each skill describes an app's structure, available actions, and handles authentication automatically.

Membrane integration skill for Outscraper data scraping APIs, enabling AI agents to extract data from Google Search, Google Maps, Amazon, YouTube, LinkedIn, and other platforms through the Membrane CLI.

An upgrade management skill from Personal AI Infrastructure (PAI), a personal AI stack framework for managing and updating your AI tool collection.

Part of a 66-skill collection for OpenCode covering 12 domains including Python, TypeScript, Go, Rust, and major frameworks like React, Django, and FastAPI. Includes 9 project workflows and 365 deep-dive reference documents loaded on demand.

Streamlines payment gateway integration with robust error handling, configuration management, and secure transaction processing scripts.

A comprehensive PDF manipulation skill for extracting text and tables, creating new PDFs, merging/splitting documents, and filling forms, supporting both Chinese and English workflows for programmatic PDF processing at scale.

An open-source cybersecurity skills library offering 754 skills across 26 security domains for AI agents, compatible with Claude Code, Cursor, Gemini CLI, and other major agent platforms.

A collection of 66 specialized skills adapted for OpenCode, spanning 12 language domains, 14 frameworks, infrastructure/DevOps, and architecture/quality categories. Includes workflows for project discovery, planning, execution, and retrospectives.

A collection of 140+ ready-to-use scientific research skills for AI agents, covering bioinformatics, cheminformatics, proteomics, clinical research, healthcare AI, medical imaging, machine learning, and materials science. Compatible with Cursor, Claude Code, and Codex via the Agent Skills standard.

A community-curated collection of custom agents, instructions, skills, hooks, workflows, and plugins for GitHub Copilot, with full-text search, MCP server tooling, and a learning hub for guides and tutorials.

An agent skill from Membrane's 3,000+ integration skills collection providing Pro Ledger accounting software access for managing accounts, contacts, invoices, and financial reports via the Membrane CLI.

Part of the AI Research Engineering Skills Library, providing AI agents with comprehensive skills for conducting AI research including model architecture, fine-tuning, data processing, and distributed training best practices.

A comprehensive collection of 140+ ready-to-use scientific skills for Claude covering bioinformatics, cheminformatics, proteomics, clinical research, healthcare AI, machine learning, and materials science, created by K-Dense.

27 first-party and 10 third-party skills for OpenCode and Claude Code, with workflow notes, design specs, and writing guidance for skill development.

A Baoyu skill that helps AI agents manage software release workflows in Claude Code, covering version management, changelog generation, and deployment coordination for improved development efficiency.

Part of the Render Skills catalog of 21 skills that teach AI coding tools how to deploy, operate, and debug apps on Render. Each skill is a self-contained SKILL.md with references, compatible with Claude Code, Codex, OpenCode, and Cursor.