feat(rag): governed reference-free scoring request privacy

[seonghobae]

Why

RAG evaluation needs a provenance-bound scoring request that cannot smuggle raw query/context/answer text.

What

• build_rag_scoring_request adapter on shared scoring contracts
• Allowlisted caller metadata; package-managed RAG provenance
• system_configuration_id must be a descriptive identifier (fail closed on raw content)
• APA/NIST doctoring + contract tests

Verification

• tests/test_scoring_rag_request.py (16)

[coderabbitai]

Warning

Review limit reached

@seonghobae, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 29 minutes

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 3627a6b8-aa48-487f-8562-d7e205d4254c

📥 Commits

Reviewing files that changed from the base of the PR and between d2301dc and dbac489.

📒 Files selected for processing (4)

• docs/changelog.d/691-rag-governed-request.md
• docs/doctoring/rag_scoring_request_privacy.md
• python/fast_mlsirm/scoring/rag.py
• tests/test_scoring_rag_request.py

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[seonghobae]

@opencode-agent address

Fresh current-main reconciliation + GREEN handoff for Draft #691. Immediately refetch exact source head, live protected main, comments/reviews and both unique target paths before writing; proceed only if source remains 1d9b7a055f778167b625c7bd2c6d43e52a136504 and live main is still compatible with 56967deceafeb7ecd0ff08a98f196282a6ec81a3. Abort if another writer appears or either ref moves.

Fresh compare is diverged, ahead 2 / behind 151, merge base a7be9eb8ca0b807364173f8471fc300e3cf9350f; unique paths are only python/fast_mlsirm/scoring/rag.py and tests/test_scoring_rag_request.py. Reconcile current protected main non-destructively first; preserve all accepted newer shared scoring/backend/security behavior and keep only this bounded RAG adapter/test slice as branch-owned work.

Then complete the already-defined privacy GREEN: validate system_configuration_id through the package shared descriptive/opaque identifier boundary before managed metadata construction; reject raw-content-shaped identity with stable non-reflective invalid_system_configuration_id; preserve valid configuration fingerprint semantics, configuration != stochastic run/respondent != generated response, caller metadata allowlist, evidence-regime/candidate-visibility semantics, query/retrieval/response fingerprints, response-size bounds and shared ScoringRequest axes. Do not add metric arithmetic, provider calls, retrieval, persistence, truth adjudication or a parallel scoring schema. Add/update authoritative RAG doctoring/changelog after GREEN and run focused/full same-head gates; keep Draft for final review/merge. Canonical Architecture/PRD/TRD/ADR/UML/ERD remains #604/#621.