{
  "updated": "2026-04-25",
  "count": 19,
  "tools": [
    {
      "name": "MCP Filesystem",
      "category": "Developer Tools",
      "description": "Read, write, move, and search files on your local machine. The best first MCP server to install — no API key, no signup, works immediately.",
      "best_for": "Anyone starting with MCP. Lets Claude read config files, write outputs, and navigate your file system without leaving the chat.",
      "website": "https://modelcontextprotocol.io",
      "github_url": "https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem",
      "api": false,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": true,
      "official": true,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "MCP Fetch",
      "category": "Developer Tools",
      "description": "Fetches any URL and returns clean, readable text — no HTML parsing needed. Lets your agent read web pages, docs, and APIs directly.",
      "best_for": "Agents that need to read live web content, check documentation, or pull data from any public URL.",
      "website": "https://modelcontextprotocol.io",
      "github_url": "https://github.com/modelcontextprotocol/servers/tree/main/src/fetch",
      "api": false,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": true,
      "official": true,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "MCP Memory",
      "category": "Developer Tools",
      "description": "Knowledge graph that persists across sessions. Your agent can store facts, relationships, and notes — and recall them in future conversations.",
      "best_for": "Agents that need to remember things between sessions. Personal assistants, research tools, anything that builds context over time.",
      "website": "https://modelcontextprotocol.io",
      "github_url": "https://github.com/modelcontextprotocol/servers/tree/main/src/memory",
      "api": false,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": true,
      "official": true,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "MCP Git",
      "category": "Developer Tools",
      "description": "Read and manipulate Git repositories — log, diff, blame, show commits, search history. Works on any local repo without touching the remote.",
      "best_for": "Developers who want Claude to understand their codebase: review changes, explain history, find when a bug was introduced.",
      "website": "https://modelcontextprotocol.io",
      "github_url": "https://github.com/modelcontextprotocol/servers/tree/main/src/git",
      "api": false,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": true,
      "official": true,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "GitHub MCP",
      "category": "Integration",
      "description": "Official GitHub MCP server, maintained by GitHub. Create issues, open pull requests, read files, search code, and manage repos — all from your agent.",
      "best_for": "Developers automating GitHub workflows: issue triage, PR summaries, code review, repo management.",
      "website": "https://github.com/github/github-mcp-server",
      "github_url": "https://github.com/github/github-mcp-server",
      "docs_url": "https://github.com/github/github-mcp-server",
      "api": true,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": true,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Playwright MCP",
      "category": "Browser Automation",
      "description": "Microsoft's MCP server for Playwright. Full browser automation — navigate pages, click, type, screenshot, extract content. Headless by default.",
      "best_for": "Agents that need to interact with web apps that don't have APIs: form submission, UI testing, structured data extraction from dynamic pages.",
      "website": "https://github.com/microsoft/playwright-mcp",
      "github_url": "https://github.com/microsoft/playwright-mcp",
      "docs_url": "https://playwright.dev/docs/intro",
      "api": false,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": true,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Figma MCP",
      "category": "Design",
      "description": "Official Figma MCP server. Lets agents read Figma files, inspect layers, extract design tokens, and retrieve component properties programmatically.",
      "best_for": "Designers and developers who want to generate code from Figma designs, or build agents that understand visual layout.",
      "website": "https://www.figma.com/developers",
      "github_url": "https://github.com/figma/mcp-server-guide",
      "docs_url": "https://developers.figma.com/docs/figma-mcp-server/",
      "api": true,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": true,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Browser Tools MCP",
      "category": "Browser Automation",
      "description": "Gives your agent eyes in the browser — take screenshots, read the DOM, click elements, monitor console logs and network requests in real time.",
      "best_for": "Vibe coders who want Claude to see and interact with what's on screen. Debugging, scraping, UI verification.",
      "website": "https://browsertools.agentdesk.ai",
      "github_url": "https://github.com/AgentDeskAI/browser-tools-mcp",
      "docs_url": "https://browsertools.agentdesk.ai",
      "api": false,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "LangChain",
      "category": "Frameworks",
      "description": "The most widely used library for building LLM applications. Chains, agents, tools, memory, and RAG — all in one framework. Python and JavaScript.",
      "best_for": "Developers who want a mature, well-documented starting point with a large community and extensive integrations.",
      "website": "https://langchain.com",
      "github_url": "https://github.com/langchain-ai/langchain",
      "docs_url": "https://python.langchain.com/docs/introduction/",
      "api": false,
      "mcp": false,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "LlamaIndex",
      "category": "Frameworks",
      "description": "Data framework for LLM applications. Specialises in indexing, retrieval, and reasoning over documents and structured data. Best-in-class RAG pipelines.",
      "best_for": "Agents that need to search, summarise, or reason over large document collections — PDFs, wikis, databases, APIs.",
      "website": "https://llamaindex.ai",
      "github_url": "https://github.com/run-llama/llama_index",
      "docs_url": "https://docs.llamaindex.ai",
      "api": false,
      "mcp": false,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "CrewAI",
      "category": "Frameworks",
      "description": "Multi-agent orchestration framework. Define a crew of agents with roles and goals — they collaborate, hand off tasks, and work in parallel to complete complex workflows.",
      "best_for": "Developers building systems where multiple agents need to work together: researcher + writer, planner + executor, reviewer + fixer.",
      "website": "https://crewai.com",
      "github_url": "https://github.com/crewAIInc/crewAI",
      "docs_url": "https://docs.crewai.com",
      "api": false,
      "mcp": false,
      "llm_agnostic": true,
      "pricing": "Free",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Langfuse",
      "category": "Observability",
      "description": "Open-source LLM observability platform. Traces every agent run end-to-end — inputs, outputs, latency, token cost, and errors. Self-hostable or cloud.",
      "best_for": "Teams who need to debug why an agent gave the wrong answer, track costs across runs, or evaluate quality over time.",
      "website": "https://langfuse.com",
      "github_url": "https://github.com/langfuse/langfuse",
      "docs_url": "https://langfuse.com/docs",
      "api": true,
      "mcp": false,
      "llm_agnostic": true,
      "pricing": "Freemium",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Helicone",
      "category": "Observability",
      "description": "LLM observability as a proxy. Route your API calls through Helicone and get instant logging, cost tracking, caching, and rate limiting — one line change, no SDK required.",
      "best_for": "Developers who want visibility into LLM usage without adding instrumentation code. Drop-in for OpenAI and Anthropic.",
      "website": "https://helicone.ai",
      "github_url": "https://github.com/Helicone/helicone",
      "docs_url": "https://docs.helicone.ai",
      "api": true,
      "mcp": false,
      "llm_agnostic": true,
      "pricing": "Freemium",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Tavily",
      "category": "Search",
      "description": "Search API designed for AI agents. Returns clean, structured results optimised for LLM consumption — no HTML, no noise, just relevant content.",
      "best_for": "Agents that need real-time web search. Works well out of the box with LangChain and LlamaIndex.",
      "website": "https://tavily.com",
      "github_url": "https://github.com/tavily-ai/tavily-python",
      "docs_url": "https://docs.tavily.com",
      "api": true,
      "mcp": false,
      "llm_agnostic": true,
      "pricing": "Freemium",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Firecrawl",
      "category": "Scraping",
      "description": "Turns any URL into clean markdown your LLM can read. Handles JavaScript rendering, pagination, authentication, and full-site crawls automatically.",
      "best_for": "Agents that need to extract content from websites — competitor research, documentation ingestion, content summarisation at scale.",
      "website": "https://firecrawl.dev",
      "github_url": "https://github.com/mendableai/firecrawl",
      "docs_url": "https://docs.firecrawl.dev",
      "api": true,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Freemium",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Pinecone",
      "category": "Vector DB",
      "description": "Managed vector database for semantic search and AI memory. Store embeddings at scale, query by similarity in milliseconds. No infrastructure to manage.",
      "best_for": "Agents that need long-term memory or RAG over large document sets. Production-ready with minimal setup.",
      "website": "https://pinecone.io",
      "github_url": "https://github.com/pinecone-io/pinecone-python-client",
      "docs_url": "https://docs.pinecone.io",
      "api": true,
      "mcp": false,
      "llm_agnostic": true,
      "pricing": "Freemium",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Qdrant",
      "category": "Vector DB",
      "description": "Open-source vector database with advanced filtering and payload search. Self-host for free or use managed cloud. Built for high-performance production workloads.",
      "best_for": "Developers who want full control over their vector store — self-hosted, open-source, no vendor lock-in.",
      "website": "https://qdrant.tech",
      "github_url": "https://github.com/qdrant/qdrant",
      "docs_url": "https://qdrant.tech/documentation/",
      "api": true,
      "mcp": false,
      "llm_agnostic": true,
      "pricing": "Freemium",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "E2B",
      "category": "Code Execution",
      "description": "Secure cloud sandboxes for running AI-generated code. Spin up an isolated environment in milliseconds, run Python or shell commands, get output back. Built specifically for agents.",
      "best_for": "Agents that write and execute code — data analysis, scripting, automation. Sandboxed so generated code can't affect your system.",
      "website": "https://e2b.dev",
      "github_url": "https://github.com/e2b-dev/e2b",
      "docs_url": "https://e2b.dev/docs",
      "api": true,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Freemium",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    },
    {
      "name": "Composio",
      "category": "Integration",
      "description": "Managed integrations platform for AI agents. 100+ apps — GitHub, Slack, Jira, Gmail, Notion — with auth handled for you. Works with LangChain, CrewAI, and direct API calls.",
      "best_for": "Agents that need to act across many tools without building OAuth flows or managing tokens for each integration.",
      "website": "https://composio.dev",
      "github_url": "https://github.com/ComposioHQ/composio",
      "docs_url": "https://docs.composio.dev",
      "api": true,
      "mcp": true,
      "llm_agnostic": true,
      "pricing": "Freemium",
      "verified": false,
      "official": false,
      "date_added": "2026-04-25",
      "submitted_by": "Svale"
    }
  ]
}