mitulgarg/env-doctor

Debug your GPU, CUDA, and AI stacks across local, Docker, and CI/CD (CLI and MCP server)

57
/ 100
Established

Performs deep stack diagnostics with compute capability matching—it detects architecture-specific mismatches like Blackwell GPUs missing sm_120 support, and validates Python version cascades across PyTorch/TensorFlow/JAX. Exposes 11+ diagnostic tools via Model Context Protocol (MCP) for integration with Claude Desktop and Zed, plus generates safe pip install commands tailored to your driver's CUDA version. Built-in CUDA auto-installer with CI/CD awareness (GitHub Actions, GitLab CI) and container validation for Dockerfiles.

125 stars and 669 monthly downloads. Available on PyPI.

Maintenance 10 / 25
Adoption 17 / 25
Maturity 22 / 25
Community 8 / 25

How are scores calculated?

Stars

125

Forks

6

Language

Python

License

MIT

Last pushed

Mar 06, 2026

Monthly downloads

669

Commits (30d)

0

Dependencies

7

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/mcp/mitulgarg/env-doctor"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.