env-doctor for AI Coding

MCP server that diagnoses local development environment issues

env-doctor, by Mitulgarg, is an open-source Model Context Protocol server that connects AI coding assistants to a developer's local environment. It lets AI agents inspect, diagnose, and suggest fixes for environment-related problems by exposing targeted tools and running on the user's machine. Key functions include scanning for missing variables and configs, verifying dependencies and runtimes, and offering automated fix steps. The tool targets developers using MCP-compatible assistants who need faster resolution of setup and 'it works on my machine' failures.

It focuses specifically on diagnosing environment setup problems

The tool acts as an MCP server that gives AI assistants executable insight into a project's runtime environment. It exposes a set of diagnostic tools an AI client can call to detect missing environment variables, absent configuration files, and mismatched runtimes. Typical outputs include:

• Missing variables and configuration files
• Dependency or runtime mismatches
• Suggested shell commands or procedural steps

Diagnosis quality depends on local visibility and needs human review

Diagnosis accuracy reflects what the local session reveals and requires operator verification. Because the tool supplies real-time local context to the assistant, suggestions match the machine's state more closely than blind remote analysis. The app generates actionable fix suggestions, but those are recommendations; users should inspect suggested commands before execution. The tool does not modify source code logic.

Input and runtime requirements constrain where it runs

The tool requires a Node.js runtime and an MCP-compatible client to operate. It supports desktop platforms where Node.js is present and integrates with clients such as Claude Desktop. Invocation via npx avoids global installation, yet the dependency on an MCP client means it cannot function without a compliant assistant connected to the local MCP endpoint. Access control is delegated to the MCP configuration.

It fits AI-assisted debugging workflows and benefits from open contribution

The tool integrates into developer workflows by supplying environment context to assistants and is maintained as open source. Lightweight invocation makes trial simple inside existing sessions. Community contributions allow inspection logic to be audited and extended. Early MCP adopters report reduced time triaging configuration issues, making the tool practical for engineers who use assistants to accelerate local troubleshooting.

Best for developers who pair AI suggestions with cautious human oversight

The tool is a pragmatic option for developers pairing MCP-capable assistants with local debugging workflows. Treat its outputs as diagnostic guidance and always inspect suggested commands before running them. For teams that enforce access controls on sensitive variables and validate AI-generated steps, the tool shortens the time spent reproducing setup errors and strengthens AI-assisted troubleshooting when used alongside human review.