events (e.g. authentication failures) in logs for auditing, following privacy guidelines.
- **Compatible with security testing:** Write code that will pass security static analysis and penetration tests (no high-severity warnings
Framework** for building and orchestrating agents
- **uv** for dependency management (NOT pip/poetry/conda)
- **pytest** for testing with pytest-mock for mocking
- **Microsoft Foundry** for infrastructure (Managed Identity, no API keys)
## Project
treat the repository
- Treat every change as a documentation edit. There is no build, test, or runtime
behaviour to reason about beyond Markdown rendering.
- Make the smallest change that achieves
file whose triggers match the task domain (git/worktree, e2e, CI, reviews, customer repro, unit tests, etc.).
- For non-trivial work also follow the chief-engineer charter's subagent workflow
flag for manual review
- **Missing header**: Automatically suggest adding the copyright header
### Exclusions
- Test files in `Tests/`, `test/`, `tests/`, or files ending with `.Tests.cs`, `.Test.cs`, `_test.py`, `test_*.py`, `.test.js`, `.spec.js
python tests/validate-instruction-fidelity.py
```
Triggered on push/PR when `manifest.yaml`, component files, or the
validation scripts change.
## Testing
There is no automated test suite. Testing uses **reference comparison**:
hand-crafted known-good prompts
user confirmation before proceeding.
- **TDD / spec-driven** — Features follow the flow: `architecture (MEDIUM/HIGH) → spec.create → test-plan → implementation → review`.
- **Least privilege** — Generate only the minimum permissions required. Use XLIFF
user confirmation before proceeding.
- **TDD / spec-driven** — Features follow the flow: `architecture (MEDIUM/HIGH) → spec.create → test-plan → implementation → review`.
- **Least privilege** — Generate only the minimum permissions required. Use XLIFF
Copilot Instructions
## Build, Test, and Lint
This is a multi-package monorepo (root, `client/`, `server/`, `scripts/`). Each has its own `node_modules` and `package.json`. The root `npm install` runs
isort directly via `script_runner.py`, providing code actions and diagnostics without the LSP server.
## Build & Test Commands
### TypeScript (extension client)
```bash
npm run compile # Webpack build (development)
npm run package # Webpack
/doc/mapping-to-hardware.md) as the authorities for language semantics; do not infer semantics from a single test or implementation detail.
- Follow [BUILDING.md](../BUILDING.md) for dependencies, out-of-tree CMake/Ninja builds, and test
Python helper scripts. There is no build system or lint config; pure-stdlib helper tests live under `test/`. The skill is consumed by future agent sessions that invoke `/hve-video-director
unresolved question, and always expand whenever callers, implementations, configuration, persistence, permissions, concurrency, retries, tests, or external boundaries can affect the behavior. A same-window 'continue' on a known active task