mlop-sl 0.0.13

mlop

git clone https://github.com/SLENSER0/ml_logger
cd ml_logger
pip install . 

import mlop

MCP server (Claude Code / Cursor)

mlop mcp runs a read-only MCP stdio server over your mlop instance, so coding agents can answer questions like "which run had the best val_acc?" or "why did eval costs spike yesterday?".

pip install 'mlop-sl[mcp]'

Register with Claude Code:

claude mcp add mlop -e MLOP_API_KEY=mlps_... -e MLOP_HOST=https://slenser0.tech -- mlop mcp

Or in a generic MCP client config (Cursor mcp.json and similar):

{
  "mcpServers": {
    "mlop": {
      "command": "mlop",
      "args": ["mcp"],
      "env": {
        "MLOP_API_KEY": "mlps_...",
        "MLOP_HOST": "https://slenser0.tech"
      }
    }
  }
}

If MLOP_API_KEY is unset, the server falls back to the token stored by mlop login. Tools: list_projects, list_runs, list_metrics, get_metric_series, search_spans, get_costs, get_run_url.

Original repo: mlop-ai/mlop