Documentation Index
Fetch the complete documentation index at: https://docs.qredence.ai/llms.txt
Use this file to discover all available pages before exploring further.
Prerequisites
- Python 3.10+
- uv package manager
- Daytona credentials for sandbox execution
- pnpm — only if you build the frontend from source
Install from PyPI
For most users, install the published package into auv-managed project:
Install from source
For contributors who need to modify fleet-rlm itself.1. Clone and sync dependencies
--all-extras installs the dev, server, and full optional extras. uv sync also installs the default dev dependency group.
2. Configure environment
.env and set at minimum:
MLFLOW_TRACKING_URI=http://127.0.0.1:5001 aligned with the make mlflow-server target. In APP_ENV=local, the API server auto-starts that local MLflow target unless MLFLOW_AUTO_START=false.
3. Frontend setup (optional)
For frontend development, install dependencies insrc/frontend/:
4. Verify the source install
Common Makefile targets
| Target | What it does |
|---|---|
make sync-all | uv sync --all-extras |
make test-fast | Run tests excluding live_llm and benchmark |
make quality-gate | Backend lint/type/tests, metadata/docs, frontend gate |
make release-check | Quality gate + security checks + UI build + packaging |
make mlflow-server | Start local MLflow server on port 5001 |
Frontend dev commands
| Command | Description |
|---|---|
pnpm run dev | Start the development server |
pnpm run api:check | Verify frontend OpenAPI artifacts match openapi.yaml |
pnpm run type-check | TypeScript type checks |
pnpm run lint:robustness | Lint lane used by repo validation |
pnpm run test:unit | Unit tests |
pnpm run build | Production build |
pnpm run check | Full suite: type-check, lint, unit, build, e2e |
Next steps
Configuration reference
Full environment variable reference.
Deployment guide
Run the API server in production.