Metadata-Version: 2.4
Name: quantifyme-mcp
Version: 0.2.0
Summary: Claude MCP server for QuantifyMe — deploy trading strategies from natural language.
Project-URL: Homepage, https://quantifyme.ai
Project-URL: Documentation, https://api.quantifyme.ai/openapi.json
Project-URL: Source, https://github.com/malcolm1232/HFTAgent
Author-email: QuantifyMe <support@quantifyme.ai>
License: MIT
Keywords: backtesting,claude,mcp,quantifyme,quantitative,trading
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Financial and Insurance Industry
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Requires-Python: >=3.10
Requires-Dist: httpx>=0.27.0
Requires-Dist: mcp>=1.0.0
Description-Content-Type: text/markdown

# quantifyme-mcp

Claude MCP server for [QuantifyMe](https://quantifyme.ai) — deploy algorithmic trading strategies from natural language.

## Install

```bash
# Recommended — isolates the install, no Python version fiddling
pipx install quantifyme-mcp
claude mcp add quantifyme quantifyme-mcp
```

Or with [uvx](https://docs.astral.sh/uv/) (no install):

```bash
claude mcp add quantifyme uvx quantifyme-mcp
```

Or plain `pip` if your default Python is 3.10+:

```bash
pip install quantifyme-mcp
claude mcp add quantifyme quantifyme-mcp
```

## Use in Claude

Once registered, ask Claude:

> Deploy a strategy that buys EURUSD when RSI hits 30 and sells when it hits 70.

Claude will call the `one_shot` tool, which generates the strategy code, trains a model on 28 days of OHLC data, and deploys it live. You get a webhook/Telegram-ready URL back.

## Environment

- `QUANTIFYME_API_KEY` — your personal key (optional; anonymous trial key auto-created if unset)
- `QUANTIFYME_BASE_URL` — override API base (default `https://api.quantifyme.ai`)

## Tools exposed

| Tool | Purpose |
|---|---|
| `one_shot` | Prompt → generate → train → deploy → live signal URL |
| `list_models` | User's trained models with train/test stats |
| `list_deployed` | Currently live models |
| `get_trial_key` | Anonymous trial key (50 credits, no signup) |
| `generate_strategy` | Code-only generation (no training) |

## License

MIT
