AI & ML

Agentic Radar is a security scanner for AI agent workflows that enables developers to visualize architectures, detect MCP servers, identify tool-based vulnerabilities, and perform adversarial testing.

A robust MCP server that integrates browser-use and Playwright to provide AI assistants with real-time web automation, deep research capabilities, and a machine-learned skill system for efficient web interaction.

A comprehensive platform for building and managing AI agents via a no-code visual builder or a TypeScript SDK with full 2-way synchronization. It supports multi-agent workflows, observability, and integration with the Model Context Protocol (MCP).

Real-time interaction with Jupyter Notebooks, allowing AI to edit, document and execute code for data analysis, visualization etc. Compatible with any Jupyter deployment (local, JupyterHub, ...).

A highly extensible and secure MCP server that allows developers to add capabilities using sandboxed WebAssembly plugins written in any language.

A laboratory and experimentation hub for integrating Azure API Management (APIM) with the Model Context Protocol (MCP) to manage AI tools, credential authorization, and security for Large Language Models.

A structured reasoning server that helps users break down complex problems into methodical, sequential thinking stages with metadata tracking and progress visualization.

A robust, modular Go framework designed for building production-grade backend applications with built-in support for observability and dependency injection.

A declarative framework that turns Markdown files and CLI tools into shareable LLM applications and MCP-compatible services. It allows users to define tools like SQL clients, grep, or curl directly in Markdown frontmatter to create modular AI agents.

Golf is a production-ready framework for building and scaling MCP servers with built-in support for enterprise authentication, OpenTelemetry, and automatic tool discovery.

A security-oriented runtime that enables AI agents to dynamically load and execute WebAssembly components as tools within a secure Wasmtime sandbox.

A unified context engineering infrastructure that provides secure, OAuth-managed MCP connections to 38+ services like GitHub, Slack, and Notion via a single endpoint.

An MCP server that enables AI agents to control web browsers using the browser-use framework, supporting OpenAI's GPT-4o for navigation and task execution.

A self-hosted registry and gateway that centralizes management for multiple MCP servers, allowing organizations to control tool access, monitor interactions, and provide a single endpoint for AI agents.

VectorCode is a repository indexing tool that uses semantic search and chunking to provide task-relevant code context to LLMs, reducing hallucinations in coding tasks.

A local knowledge graph system providing persistent memory for AI models through structured entities, relations, and context-aware storage. It supports both global and project-local databases with built-in safety markers to ensure secure data management.

An MCP server for codebase indexing and semantic search that provides automatic incremental updates, .gitignore integration, and multi-encoding support to find relevant code snippets.

A macOS-specific MCP server that bridges Claude with the ChatGPT desktop application via AppleScript, allowing for cross-assistant communication and conversation management.

A specialized MCP server that deploys autonomous Browser-Use agents to perform multi-agent QA testing on local or live websites for UI, accessibility, and technical defects.

connects QGIS to Claude AI through the MCP. This integration enables prompt-assisted project creation, layer loading, code execution, and more.

A comprehensive MCP server that enables natural language management of Kubernetes clusters, supporting pods, deployments, Helm charts, and security auditing.

k8m is an AI-powered lightweight Kubernetes dashboard and MCP server that enables large language models to manage clusters via 49 built-in tools.

Debug your AI agents

`Unity Editor` and `Unity Runtime` MCP integration. Unit Test, Coding, C# Roslyn, Reflection, Assets. Helps to create games with AI. And helps to run AI logic in the game in runtime.

Make your WordPress site into a simple MCP server, exposing functionality to LLMs and AI agents.

Context Portal (ConPort) is a memory bank database system that effectively builds a project-specific knowledge graph, capturing entities like decisions, progress, and architecture, along with their relationships. This serves as a powerful backend for Retrieval Augmented Generation (RAG), enabling AI assistants to access precise, up-to-date project information.

An MCP server that enables web search capabilities using DuckDuckGo and provides intelligent content fetching and parsing for LLM-friendly consumption.

A powerful Model Context Protocol (MCP) server that provides seamless access to Apple Developer Documentation through natural language queries. Search, explore, and get detailed information about Apple frameworks, APIs, sample code, and more directly in your AI-powered development environment.

An all-in-one developer tool and SDK designed for debugging, managing, and deploying Model Context Protocol (MCP) servers with integrated inspector and interactive testing capabilities.

Hyperbrowser is the next-generation platform empowering AI agents and enabling effortless, scalable browser automation.