Testing and Quality
The repo already tracks quality at the domain level, which is useful because the codebase spans APIs, realtime handlers, content systems, connectors, and infrastructure logic.
Core Commands
uv run pytest
uv run ruff check path/to/changed.py
uv run ruff format path/to/changed.py
cd docs/docusaurus && npm run build
cd frontend && npm run lint
Test Surface
The src/tests/ tree includes:
- API tests
- integration tests
- repository tests
- unit tests
This is important because some regressions only show up when the full request/event stack is exercised.
Current Quality Signals
The root quality score document highlights:
- strong coverage in config and billing-heavy areas
- mid-range confidence in agent/chat orchestration surfaces
- weaker documentation and test maturity in content and some integrations
Top Improvement Priorities
- content domains such as slides, storybooks, and media
- integrations and connector-related areas
- agent socket orchestration
- deployment reliability and documentation
Release Discipline
At minimum:
- run tests relevant to the change
- run Ruff on changed Python files
- build docs when docs changed
- validate the service mode you actually touched
Large repos fail when people only test the slice they edited and ignore the runtime mode that exposes it.