Rust And Benchmark Hygiene Audit¶

Scope¶

This audit covers repository-hygiene follow-up for two recurring drift points:

the Rust wrapper lockfile policy around rust-cli/Cargo.lock
the benchmark corpus filter used by notebook and training assertions as audit notes and samples/logs/ continue to grow

REPO_COMPLETENESS_CHECKLIST.md now states that rust-cli/Cargo.lock is committed intentionally so the Rust wrapper behaves like an application-style entrypoint with reproducible dependency resolution.
TODO.MD now treats the Rust lockfile decision as established policy instead of an open question.
src/repo_rag_lab/benchmarks.py now exposes is_benchmark_document_path(...) and excludes docs/audit/ in addition to the existing generated and operational surfaces.
benchmark and project-surface tests now pin the benchmark exclusion policy and the committed Rust lockfile policy.

Executed successfully in this turn:

uv run python -m compileall src tests
uv run pytest tests/test_benchmarks_and_notebook_scaffolding.py tests/test_project_surfaces.py
cargo build --manifest-path rust-cli/Cargo.toml

Benchmark assertions now ignore docs/audit/ alongside samples/logs/, samples/training/, samples/population/, artifacts/, and other non-corpus surfaces that would otherwise let retrieval tests pass against operational summaries instead of the intended repository content.
The repository now has a documented answer for rust-cli/Cargo.lock: keep it committed and refresh it deliberately when Rust dependencies change.