Arize-ai/phoenix
Phoenix is an open-source AI observability platform for experimentation, evaluation, and troubleshooting of large language model applications, supporting multiple frameworks and LLM providers with flexible deployment options.
Awesome MCP › Testing & Debugging
hyprmcp/jetski
Jetski is an open source MCP analytics and authentication platform from HyprMCP, available as a self hosted install or as the managed HyprMCP Cloud service. It targets three problems teams hit after publishing an MCP server: users who never finish setup because client instructions are hard to find or because they open the MCP URL in a browser, authentication that is painful to implement, and the absence of any visibility into how the server is used. Jetski addresses these without changes to the MCP server itself. The platform manages and deploys instances of the separate mcp-gateway component, an OAuth proxy with dynamic client registration that sits in front of the target MCP server, and it collects and aggregates the analytics and metrics that the gateway captures. The dashboard provides prompt analytics that link which prompt triggered which tool from which client and what the result was, automatically generated setup instructions for every supported MCP client through the companion mcp-install-instructions-generator project, zero code authentication so users can be identified and served personalised tools and responses, and real time debug logs of each request, response and error. The development setup uses mise to provision the toolchain, pnpm for the frontend, Docker Compose for supporting services including a Dex identity provider, and a local hosts file entry; backend and frontend run as two processes and a generator task seeds demo data before logging in with the sample admin account. Optional Kubernetes orchestration of gateway instances is tested with Minikube and installs a metacontroller into the current kubectl context. The project states that it is under active development and that both the API and the command line interface may change in backwards incompatible ways. It fits teams operating MCP servers that need usage insight and access control.
https://github.com/hyprmcp/jetski
Also in Testing & Debugging
modelcontextprotocol/inspector
MCP Inspector is a developer tool that provides a visual UI and CLI for testing, debugging, and interacting with Model Context Protocol (MCP) servers, enhancing MCP server development workflows.
MCPJam/inspector
MCPJam Inspector is a developer tool for testing and debugging MCP servers, supporting multiple protocols and LLM interaction, designed to streamline MCP development workflows.
Arize-ai/openinference
OpenInference is an open-source extension of OpenTelemetry that provides comprehensive tracing and observability for AI applications, including support for the Model Context Protocol (MCP).
LSTM-Kirigaya/openmcp-client
OpenMCP is an all-in-one VSCode plugin that integrates development, testing, and management tools for Model Context Protocol (MCP) server debugging and large model interaction.
workbackai/mcp-nodejs-debugger
MCP Node.js Debugger is an MCP server that enables AI coding assistants like Cursor and Claude Code to debug Node.js applications at runtime by setting breakpoints and inspecting runtime state.
modelscope/MCPBench
MCPBench is an open-source evaluation framework for benchmarking Model Context Protocol (MCP) servers across Web Search, Database Query, and GAIA tasks, measuring accuracy, latency, and token consumption under consistent LLM and agent settings.
loopwork/Companion
Swift app for macOS, iOS and visionOS that connects to local and remote MCP servers to browse and exercise their prompts, resources and tools while testing and debugging.