inspector: Test and Debug MCP Servers and Apps

Summary
MCPJam Inspector helps developers test, inspect, and evaluate MCP servers and apps across clients and models. Use it to debug protocol and OAuth behavior, run repeatable evaluations, and check for regressions in CI.
At a glance
- Language
- TypeScript
- License
- NOASSERTION
- Stars
- 2.2k
- Forks
- 293
- Added to OSRepos
- March 5, 2026
- Last analyzed
- October 3, 2026
Topics
Click on any tag to explore related repositories
Use at your own risk
OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.
Overview
MCPJam Inspector is a testing and evaluation platform for developers building MCP servers, MCP apps, and ChatGPT apps. It brings server interactions, client behavior, protocol traces, and model-driven tests into one workflow, helping identify integration differences and failures before deployment.
It is useful both for hands-on local debugging and repeatable checks in automated pipelines. The project provides a hosted app, desktop app, terminal interface, and SDK, with local options for connecting to HTTP/S and STDIO servers.
Key Features
- Inspect and invoke tools, resources, resource templates, prompts, and elicitation flows with JSON-RPC logs.
- Exercise servers through a chat playground with trace and raw views, including support for MCP app and OpenAI Apps SDK interfaces.
- Debug OAuth flows and run guided protocol conformance checks.
- Define test cases with expected tool calls, run evaluations across models, and track accuracy over time.
- Compare behavior across client configurations and inspect traces of agent steps and tool calls.
- Use a CLI and SDK for server checks, scripted inspection, and CI/CD workflows.
- Share server configurations through team workspaces with real-time sync.
Use Cases
- An MCP server developer can reproduce a tool or resource failure and inspect the underlying JSON-RPC messages during local development.
- A team integrating with multiple AI clients can compare behavior across client configurations before releasing an integration.
- An engineer responsible for authentication can diagnose OAuth errors and check conformance to supported protocol versions.
- A team can run evaluation cases and conformance checks in CI to catch behavior changes before merging or deploying.
Project Facts
- Language: TypeScript
- License: NOASSERTION
- Stars: 2.2k
- Forks: 293
- Topics: anthropic, chatgpt, cicd, debugger, evals, evaluation, inspector, mcp, mcp-apps, mcp-clients, mcp-inspector, mcp-server, mcp-tools, modelcontextprotocol, oauth, oauth2, openai, openai-apps-sdk, opensource, tracing
- Archived: no
Getting Started
Run the local terminal version with Node.js 20 or later:
npx @mcpjam/inspector@latest
Alternatively, open the hosted app. See the repository README for desktop, Docker, and configuration details.
Alternatives
- promptfoo: Promptfoo evaluates prompts and LLM applications, including security checks, while Inspector focuses on testing MCP servers and apps across clients and models.
Considerations
The hosted app supports HTTPS server URLs only. Local STDIO connections and some local workflows require the terminal or desktop app. The terminal install requires Node.js 20 or later. The repository metadata reports the license as NOASSERTION, while the README states that the project is licensed under Apache License 2.0. The project is active and has a substantial open issue count, so check the issue tracker for current reports relevant to your setup.
Source repository
Open the original repository on GitHub.
15 counted GitHub visits
Related repositories
Similar repositories that may be relevant next.

OrcaReplay: Record and Replay AI Agent Runs
October 2, 2026
OrcaReplay records AI agent runs so developers can inspect what happened, replay interactions offline, and fork a run from a checkpoint onto another model. It is a TypeScript CLI for debugging and comparing agent behavior without modifying the agent.

openmake_llm: Coordinate AI Models, Agents, and Tools
October 1, 2026
OpenMake is a self-hosted AI workspace that coordinates chat models, agents, and tools in one place. It suits people who want multi-step AI workflows with local or open-weight models, while keeping infrastructure and model choices under their control.

zennotes: Manage Markdown Notes with Keyboard-First Tools
October 1, 2026
ZenNotes is a local-first Markdown notes app for people who prefer plain files, keyboard navigation, and Vim motions. It runs as an Electron desktop app or a self-hosted web app, with search, diagrams, CLI, and MCP integrations.

Kun: Local-First AI Agent Workspace
September 27, 2026
Kun is a local-first workspace for using AI agents across coding, writing, design, research, and automation. Its desktop GUI and terminal TUI share one runtime, keeping tasks and approvals connected.