Metadata-Version: 2.4
Name: autooptm
Version: 0.1.0
Summary: Client, CLI and MCP server for AutoOptm — measured GPU speedups for your training and inference code
License: MIT
Requires-Python: >=3.9
Provides-Extra: mcp
Requires-Dist: mcp>=1.0; extra == 'mcp'
Description-Content-Type: text/markdown

# autooptm

Measured GPU speedups for your training and inference code — from the
terminal, an agent, or Python. Diagnosis and measurement are free; credits
move only when you unlock the patch.

```bash
pip install autooptm
autooptm login                 # email code, no browser needed
autooptm run https://github.com/you/repo --entrypoint train.py
autooptm unlock <jobId>        # asks before spending credits
git apply autooptm.patch
```

Python:

```python
from autooptm import AutoOptm
ao = AutoOptm()
job = ao.submit(".", entrypoint="bench.py", workload="inference")
done = ao.wait(job["jobId"])
print(done["speedup"], done["unlock_credits"])
```

MCP (Claude Code, Cursor, any MCP client) — `pip install 'autooptm[mcp]'`:

```json
{ "mcpServers": { "autooptm": { "command": "autooptm-mcp" } } }
```

Tools: `optimize_submit`, `optimize_status`, `optimize_wait`,
`unlock_patch` (requires explicit user confirmation), `download_patch`,
`account_balance`.
