Explore the directory

A curated list of servers, clients, frameworks, and tools implementing Google's Agent2Agent protocol for interoperable AI agent systems.

• Curated lists

An MCP server that connects Claude, Codex, and other MCP clients to a live Abaqus/CAE session, so an AI assistant can build and run finite-element models from a description.

• MCP servers

A self-hosted, ChatGPT-style web interface for vLLM and Ollama, built with Next.js, that runs entirely in the browser with chats stored in local storage.

• Inference

A Zotero plugin that adds local, offline semantic search over a research library, plus a built-in read-only MCP server so AI coding agents can search and cite the same papers.

• MCP servers
• RAG

Three small command-line tools that let LLM coding assistants evaluate Clojure code in a live REPL and automatically fix mismatched delimiters.

• CLI tools
• Debugging

A reference architecture demonstrating a multi-agent chatbot built with Strands Agents and Amazon Bedrock AgentCore, combining tool execution, memory, and browser automation.

• Agent frameworks
• Tutorials

An R package for building retrieval-augmented generation pipelines, with sensible defaults for chunking, embedding, and storage while keeping every step inspectable.

• RAG
• Libraries

A schema-based extractor that pulls structured data from documents and the web into JSON or Markdown with a single tool call, for use inside AI agent workflows.

• RAG
• Data engineering

A macOS menu bar app that exposes Apple's on-device Foundation Models through OpenAI- and Ollama-compatible local API endpoints, so nothing leaves the machine.

• Inference
• Desktop utilities

A native macOS app for running large language models locally on Intel Macs with AMD GPUs, using a patched llama.cpp build to fix corrupted output and low speed on that hardware.

• Inference

A Java framework for building agentic AI applications that turn natural-language prompts into calls against enterprise Java methods, REST APIs, or shell scripts.

• Agent frameworks
• Libraries

The official MCP server for StarRocks, letting an AI assistant run SQL, explore schemas, and chart query results against a StarRocks database.

• MCP servers
• Databases

An experimental MCP server from Mozilla that gives LLMs and coding agents access to MDN's search, documentation, and browser compatibility data.

• MCP servers
• Documentation tools

An MCP server that gives Claude, Cursor, and other MCP clients access to live tennis scores, rankings, fixtures, and a historical match archive dating back to 1968.

• MCP servers
• APIs

An open-source Chrome extension that reads a page's DOM to plan and carry out browser automation from a plain-English instruction, for interactive use and MCP clients.

• Browser automation

A self-hosted RAG document assistant with per-user knowledge bases, PDF indexing, source citations, and a configurable distance threshold that makes it decline out-of-scope questions.

• RAG

A distributed, peer-to-peer system for publishing, discovering, and verifying metadata about AI agents by their capabilities and dependencies.

• Agent frameworks

A framework-agnostic, cloud-agnostic runtime for running enterprise AI agents in production, with queued execution, built-in guardrails, and shared observability.

• Agent frameworks
• Hosting

A research implementation of a transformer architecture for 3D scene understanding, partitioning a scene into volumetric patches for semantic and instance segmentation.

• Papers

An MCP server that gives AI assistants access to TikTok video subtitles, post details, and search results through a third-party API.

• MCP servers
• APIs

A native Svelte 5 data grid and table, usable as a drop-in render component or as a standalone headless state machine, with virtualization for very large datasets.

• Libraries
• UI kits

A multi-agent AI coding assistant aimed at Chinese developers, coordinating dozens of specialized agents on top of DeepSeek, GLM, and other domestic language models.

• Coding agents

A Go framework for building agentic LLM applications, with a common interface across model providers, MCP tool support, and automatic session and memory management.

• Agent frameworks
• Libraries

A minimal terminal and browser-based coding and task assistant that talks to OpenAI-compatible and Anthropic APIs, keeping sessions, skills, and memory on the user's machine.

• Coding agents
• Terminals