Unless the user explicitly asks otherwise, treat the following subsystems as out of scope for analysis, planning, implementation, refactoring, testing, and documentation work:
- TTAS and tiered tensor memory
- Queue and its samplers
- Search
- the standalone
persisting-dlcapt/dlcaptcomponent
Do not modify these subsystems, expand their APIs, fix their tests, or include their failures in the acceptance criteria for an unrelated task. Prefer targeted build, lint, and test commands over workspace-wide commands when a workspace-wide command would pull them into scope.
This exclusion covers, in particular:
- TTAS and tiered-memory code
persisting/queue/,persisting/sampler/, and their tests and docspersisting/search/,crates/persisting-pchronicle/src/search/, Search CLI surfaces, and their tests and docscrates/persisting-dlcapt/and dlcapt-specific scripts, tests, features, and workflows
The default active scope is the Agent infrastructure centered on pVisor,
pPilot, pChronicle, Gateway, Control, OverlayFS, OverlayNet, and trajectory CLI
surfaces. persisting-dlcapt is a separate standalone component; excluding it
does not exclude Gateway trajectory capture or pChronicle capture storage.
Enter an excluded subsystem only when:
- the user explicitly names that subsystem in the current task; or
- an in-scope change cannot be completed without a minimal dependency-boundary adjustment there.
For the second case, keep the adjustment minimal and explain why it is required. Do not use incidental cleanup as a reason to broaden scope.
Use just test as the default validation command. It runs Rust tests through
cargo nextest in release mode and then runs the Python test suite. To limit
Rust coverage to one package, pass its Cargo package name positionally, for
example just test persisting-agentctl. Use a direct cargo test invocation
only when doctests or an explicitly documented special runner is required.