Skills directory
Skills are abilities you can add to an AI assistant — browsing the web, reading your files, searching Notion. Curated, with one-click install for Claude Desktop, Claude Code, Cursor, and VS Code.
New to this? Use your first skill in 5 minutes → · Confused by the terms? Agents vs. skills explained →
Browser automation
These skills let an AI assistant drive a real web browser — open pages, click, fill forms, take screenshots. Useful for price comparison, form filling, and anything a human would do by hand in a browser.
Web search & scraping
Give your assistant live access to the web: search engines, page fetching, and structured scraping. Without one of these, most assistants only know what they were trained on.
Live web search results inside your assistant.
Fetches a single web page and converts it to readable text.
Industrial-strength web scraping with JavaScript rendering.
Semantic web search built for AI assistants.
Search + extract, tuned for AI research workflows.
Files on your computer
Let the assistant read, search, and organize files in folders you explicitly allow — PDFs, spreadsheets, documents. Nothing leaves your machine except what you ask about.
Productivity apps
Connect the tools you already work in — Notion, calendars, spreadsheets — so the assistant can search and update them for you.
Project management
Linear, Asana, Jira & Co. — let the assistant read ticket status, draft issues, and summarize what your team shipped. Most of these are hosted by the vendor: zero install, browser sign-in only.
Developer tools
Skills for people who code: repository access, up-to-date library documentation, version control, structured reasoning.
Issues, pull requests, and code from your GitHub account.
Up-to-date documentation for any library, on demand.
Read, search, and analyze local Git repositories.
Structured step-by-step reasoning for complex problems.
Databases
Query your own databases in plain language. Start with read-only credentials — these skills execute real queries against real data.
Hosting & infrastructure
Deployments, DNS, error tracking — ask your assistant what broke and where, instead of clicking through dashboards.
Design & web building
Connect design and site-building tools so the assistant can read structures and make changes where the vendor allows it.
Communication & support
Customer conversations and support data — summarize tickets, draft replies, spot patterns.
Knowledge & research
Skills that make the assistant smarter about external knowledge: code-base wikis, AI model hubs, structured research sources. Several are completely open — no account at all.
Audio, image & video
Generate and process media from inside your assistant — voice-overs, transcriptions, sound effects.
Payments & business
Query payment and business data from inside your assistant. Read-oriented usage recommended — keep write access behind test keys until you trust the workflow.
Memory
By default assistants forget everything between conversations. Memory skills give them a persistent notebook so preferences and facts survive across chats.
Building for agents? Check your own domain.
agentics audits whether AI agents can read, find, and trust your domain — robots.txt, llms.txt, JSON-LD, MCP discovery, and email deliverability in one free check.
Run free check →