# a2a mcp > A2A MCP is a directory and editorial resource for Model Context Protocol (MCP) servers, Agent2Agent (A2A) agents, AI agents, frameworks, clients, and developer tools. Official website: https://a2a-mcp.org/ Primary language: English. Translated listings and articles use their own canonical slugs and identify their language in the Markdown document. The site helps developers discover projects and learn about agent interoperability. Listings describe projects maintained by their respective providers; A2A MCP is not the official specification publisher for MCP or A2A. Use the Markdown links for page content without navigation, advertising, or JavaScript. Each document includes its canonical HTML URL and update date. Follow the project's website and cited sources for current capabilities and setup instructions. ## Core pages - [Home](https://a2a-mcp.org/): Browse the directory and featured content. - [Latest entries](https://a2a-mcp.org/latest): Recently added tools and resources. - [Blog](https://a2a-mcp.org/blog): Articles, tutorials, and protocol news. - [Agent skills](https://a2a-mcp.org/skills): Discover reusable agent skills. - [A2A vs MCP](https://a2a-mcp.org/compare/a2a-vs-mcp): Compare the protocols and their roles. - [A2A Agent Card Validator](https://a2a-mcp.org/tools/a2a-agent-card-validator): Validate an A2A agent card in the browser. ## Categories - [MCP Servers](https://a2a-mcp.org/category/mcp-server): MCP-compatible servers providing tools, data, and API integrations for AI models. - [A2A Agents](https://a2a-mcp.org/category/a2a-agent): Agents using the A2A protocol for cross-vendor collaboration and task delegation. - [AI Agents](https://a2a-mcp.org/category/ai-agent): Autonomous AI agents for coding, browsing, research, and workflow automation. - [Frameworks & SDKs](https://a2a-mcp.org/category/framework): SDKs, libraries, and frameworks for building agents and protocol integrations. - [Tools & Extensions](https://a2a-mcp.org/category/tools): Developer tools, extensions, and ecosystem utilities. - [Getting Started](https://a2a-mcp.org/category/getting-started): Step-by-step guides, installation tutorials, and quickstart resources. - [Deep Dives](https://a2a-mcp.org/category/deep-dives): Architectural analysis, protocol comparisons, and technical research. - [Video Tutorials](https://a2a-mcp.org/category/videos): Practical demonstrations and video tutorials. - [Use Cases](https://a2a-mcp.org/category/use-cases): Real-world applications, automation cases, and multi-agent orchestration. - [Clients & Apps](https://a2a-mcp.org/category/client): Desktop, mobile, and web clients for interacting with AI agents. - [Security & Risks](https://a2a-mcp.org/category/security): Vulnerability alerts, supply chain defense, and security best practices. ## Selected entries - [Weibo AI Bridge](https://a2a-mcp.org/entry/weibo-ai-bridge.md): Weibo AI Bridge routes Weibo private messages to local agent CLIs and streams their responses back to the conversation. - [OpenCLI](https://a2a-mcp.org/entry/opencli.md): OpenCLI gives AI agents deterministic commands for websites, browser sessions and local tools, including documented Weibo adapters. - [Agent Client Protocol](https://a2a-mcp.org/entry/agent-client-protocol.md): Agent Client Protocol standardizes communication between coding agents and client applications such as code editors. - [Vercel AI SDK](https://a2a-mcp.org/entry/vercel-ai-sdk.md): Vercel AI SDK is an open-source TypeScript toolkit for building AI-powered applications and agents. - [OpenAI Agents SDK](https://a2a-mcp.org/entry/openai-agents-sdk.md): OpenAI Agents SDK provides application-level building blocks for agent workflows in Python and TypeScript. - [Microsoft AutoGen](https://a2a-mcp.org/entry/microsoft-autogen.md): AutoGen is a programming framework for agentic applications and multi-agent workflows, retained for teams evaluating or maintaining existing implementations. - [Mastra](https://a2a-mcp.org/entry/mastra.md): Mastra is a TypeScript framework for building AI agents and applications with tools, memory and workflow capabilities. - [LangGraph](https://a2a-mcp.org/entry/langgraph.md): LangGraph is an orchestration framework and runtime for building long-running, stateful agents and controlled agent workflows. - [CrewAI](https://a2a-mcp.org/entry/crewai.md): CrewAI is a Python framework for building collaborative agents, role-based crews and controlled workflow flows. - [Luma Agents](https://a2a-mcp.org/entry/luma-agents.md): Luma Agents coordinate creative work across image, video and other media, carrying a project from a brief through iteration and delivery. - [Zed Agent](https://a2a-mcp.org/entry/zed-agent.md): Zed Agent is the editor’s native coding agent, combining project context, file editing, terminal tools and Zed-managed MCP integrations. - [Pi Coding Agent](https://a2a-mcp.org/entry/pi-coding-agent.md): Pi is a terminal coding harness with file and shell tools, persistent sessions and extensions for customizing agent workflows. - [Cline](https://a2a-mcp.org/entry/cline.md): Cline is an AI coding agent that works with project files and development tools and can extend its capabilities through MCP servers. - [GitHub Copilot Agent Mode in VS Code](https://a2a-mcp.org/entry/github-copilot-agent-mode.md): GitHub Copilot’s agent workflow in VS Code combines project context, file edits, terminal tools and configured MCP tools to carry out coding tasks. - [Cursor Cloud Agents](https://a2a-mcp.org/entry/cursor-background-agent.md): Cursor Cloud Agents run development tasks in remote environments, handling code changes and reporting results back for review. - [Devin](https://a2a-mcp.org/entry/devin.md): Devin is an AI software engineering agent that can work on code, run tests and carry out scoped development tasks. ## Recently updated articles - [Javii Tools Review: The Tiny Browser Suite Behind Pixel-Perfect Creator Visuals](https://a2a-mcp.org/blog/javii-tools-review.md): A practical Javii Tools review covering features, pricing, privacy, licensing, alternatives, and whether it fits creator workflows. - [Claude Response Incomplete? The Real Causes, Fixes, and Prevention Checklist](https://a2a-mcp.org/blog/claude-response-incomplete.md): Fix incomplete Claude responses with token, context, tool-use, outage, and prompt troubleshooting steps for Claude.ai and API users. - [Gemini Error 1076: What It Means, Why It Happens, and How to Fix It Fast](https://a2a-mcp.org/blog/gemini-error-1076.md): Fix Gemini error 1076 with outage checks, account tests, chat recovery, browser fixes, and API troubleshooting steps. - [Anthropic’s 1GW Data Center Push: Why AI Companies Are Now Fighting for Power, Not Just Models](https://a2a-mcp.org/blog/anthropic-1gw-data-center-leases-ai-infrastructure.md): Anthropic’s 1GW data center lease plan shows how AI competition is shifting from models to power, chips, financing, and infrastructure. - [Codex Plus to Pro Still Hit Usage Limit? The CLI Re-Login Fix Explained](https://a2a-mcp.org/blog/codex-plus-to-pro-usage-limit-cli-login-fix.md): Upgraded from Codex Plus to Pro but still hit usage limits? Learn why Codex CLI may need logout and login to refresh Pro access. - [How to Use Claude Fable 5: Complete 2026 Guide to Anthropic’s Most Powerful Public AI Model](https://a2a-mcp.org/blog/how-to-use-claude-fable-5.md): Master Claude Fable 5 with this definitive guide. Learn access methods, advanced prompting, API setup, safety handling, and pro tips for long-horizon coding and agentic tasks from Anthropic’s latest model. - [Fable 5 Subscription Guide: Plans, Usage Credits, Pricing, Limits, and When It’s Worth Paying For](https://a2a-mcp.org/blog/fable-5-subscription.md): Understand Fable 5 subscription access, credits, API pricing, limits, retention rules, and whether it is worth upgrading. - [How to Change Claude Code to Fable 5: Step-by-Step Model Switching Tutorial](https://a2a-mcp.org/blog/claude-code-change-model-to-fable-5.md): Learn how to switch Claude Code to Fable 5 using /model, CLI flags, settings, and environment variables with troubleshooting tips. - [Claude Fable 5 vs OpenAI GPT-5.5: Which Frontier AI Model Fits Your Workflow?](https://a2a-mcp.org/blog/claude-fable-5-vs-openai-gpt-5-5.md): Compare Claude Fable 5 and OpenAI GPT-5.5 across pricing, context, coding, tools, reasoning, API access, and real-world use cases. - [How to Access the Fable 5 API: A Developer’s Guide to Claude’s Mythos-Class Model](https://a2a-mcp.org/blog/how-to-access-the-fable-5-api.md): Learn how to access the Fable 5 API, use the correct model IDs, compare pricing, and choose Anthropic, Bedrock, or Vertex AI. - [Claude Fable 5 Cost Explained: Pricing, Trade-Offs, and Real-World API Scenarios](https://a2a-mcp.org/blog/how-claude-fable-5-cost.md): Compare Claude Fable 5 costs by input, output, caching, US-only inference, and real-world API scenarios. - [Fable 5 Explained: The Mythos-Class Claude Model Built for Long-Horizon AI Work](https://a2a-mcp.org/blog/fable-5.md): A complete guide to Fable 5: capabilities, pricing, use cases, safeguards, developer tips, and how it compares with Mythos 5. ## About - [About A2A MCP](https://a2a-mcp.org/about): Site purpose and editorial approach. - [Contact](https://a2a-mcp.org/contact): Contact the site about listings and content. ## Optional - [Sitemap](https://a2a-mcp.org/sitemap.xml): Full index of public HTML pages. - [Privacy policy](https://a2a-mcp.org/privacy) - [Terms](https://a2a-mcp.org/terms)