inspectr
Inspectr – easily view and analyze API requests, MCP server calls and webhook events.
claude mcp add --transport stdio inspectr-hq-inspectr npx -y @inspectr/inspectr
How to use
Inspectr is a lightweight API proxy and MCP server designed to help you debug API calls, MCP tool interactions, and webhook events. The server exposes a convenient UI and tooling for analyzing requests, replaying captured traffic, and exposing local services publicly. With the provided MCP integration, you can run Inspectr via npm’s npx launcher and start using its MCP endpoints to view and manage tool calls, prompts, and resources used by LLMs interacting with your MCP server. The built-in tunnel/exposure features also let you securely share a local MCP server to external systems for testing and collaboration. To get started, launch Inspectr with the npx command and point it at your backend as shown in the quick start: npx @inspectr/inspectr --backend=http://localhost:3000, then open the Inspectr UI to inspect requests, webhooks, and MCP activity in real time.
How to install
Prerequisites:
- Node.js (v14+ recommended) and npm installed on your development machine
- Basic familiarity with running CLI commands
Installation steps:
- Install Node.js from https://nodejs.org/ if you don’t have it already.
- Use npx to run Inspectr directly without a global install (the MCP entrypoint):
# Run Inspectr (no local install required)
npx @inspectr/inspectr --backend=http://localhost:3000
- If you prefer to install locally first, you can install the package (optional) and then run it via node/npm scripts:
# Optionally install locally (not required for npx usage)
npm install @inspectr/inspectr
# Run (example, if a local executable is provided by the package)
npx @inspectr/inspectr --backend=http://localhost:3000
- Start the MCP server and access the UI at the specified port (default UI host/port as per documentation). Adjust the backend URL to match your API server. You can also refer to the Inspectr docs for advanced configuration options such as tunneling/public exposure and MCP-specific settings.
Additional notes
Tips and notes:
- The MCP server is intended to work with Node-based deployments via npx. If you need a different deployment method, consider using the recommended mcp_config entry with the appropriate command type.
- Ensure your backend (the API you’re proxying or debugging) is reachable from the Inspectr server. The --backend flag tells Inspectr where to forward traffic for analysis.
- If you plan to expose a local service publicly, configure the tunnel/subdomain options per the Inspectr docs to maintain security and control visibility.
- For MCP-related tooling, Inspectr’s UI provides insights into prompts, resources, and token usage when interacting with LLMs, which can help diagnose performance or usage issues.
- Check for environment variable requirements in the official docs if you’re deploying in containers or CI environments.
Related MCP Servers
anytype
An MCP server enabling AI assistants to interact with Anytype - your encrypted, local and collaborative wiki - to organize objects, lists, and more through natural language.
flapi
API Framework heavily relying on the power of DuckDB and DuckDB extensions. Ready to build performant and cost-efficient APIs on top of BigQuery or Snowflake for AI Agents and Data Apps
storyblok
A modular, extensible MCP Server for managing Storyblok spaces, stories, components, assets, workflows, and more via the Model Context Protocol (MCP).
openproject
A Model Context Protocol (MCP) server that provides seamless integration with OpenProject API v3.
appdog
Compose and generate effortlessly MCP servers from any OpenAPI specifications
mcp
Model Context Protocol (MCP) server for video automation with Plainly Videos API (Adobe After Effects automation).