governance(v21): authorize Learned Policy V3 production closure

[laiqian0239-glitch]

Fresh-main successor authorization for Learned Policy P1 production closure.

Base: a45b11db14330a64e61f933d20d92b61bbcd9d9e
Exact current Authorization Head: b2eb6c1c38b6c0cbafea74cc774fa13f94799673
Initial proposal Head / causal authorization RED: f0cfbc0fa7add74cc383355bce8f383bb0acfa90
Initial Stage RED: run 31840999804, job 94897691232, exact cause: missing required top-level OSS-first admission fields under ossFit.
Root-fix descendant: b2eb6c1c38b6c0cbafea74cc774fa13f94799673, adding only ossFit.decision, matureOssAvailable, selectedAdoptionMode, and newGeneralPurposeInfrastructure; implementation scope/digests are unchanged.

This proposal supersedes the V2 Learned Policy authorization/implementation authority after ordinary merge. PR #379 remains immutable audit evidence and must not be merged, rebased, amended, force-pushed, or reused as executable authority.

The V3 implementation seal is exactly 28 paths with SHA-256 0c956d8e55cdfdf51996de3d80cd8e2df32e9b13e5323a823419e81e8c9b21bb. The mandatory first implementation commit remains exactly the six frozen Learned Policy tests with SHA-256 9aff346a55b16f1ae54a743daf15bb18cffc4c1471581556be1eea22acab3142, producing a fresh causal RED from the V3 authorization merge parent. The first post-RED production commit must bind the real RED Head/run/conclusion trailers.

Root closure adds no new general-purpose Yance infrastructure: it reuses existing OpenFeature + flagd in-process offline native JSON for explicit Promotion/Rollback activation, existing Learning promotion authority, Vowpal Wabbit 9.11.2 in the sealed Learning runtime, and the existing WP7 presealed-runtime trusted-product pattern. Production-default composition must resolve a promoted content-addressed policy and consume its bounded candidateStrategyBranch before existing Model Brain/LiteLLM frontier generation. UAT-only injected VW execution does not count as production closure.

No root npm manifest/lock changes, no policy store/rollout database, no flagd daemon, no new model gateway/reward engine, no randomized live exploration, no formal release/publish/promotion claim.

Summary by CodeRabbit

• New Features

  • Added a V3 production-closure authorization document for the V21 Learning Policy workflow.
  • Defined deterministic policy handling, content-addressed artifacts, sealed Windows runtime requirements, and implementation controls.
  • Added mandatory testing, governance states, and authorization criteria.

• Documentation

  • Recorded prior failure evidence and clarified prohibited approaches, including runtime dependency resolution, request overrides, duplicate infrastructure, and randomized exploration.
  • Clarified that production, release, publishing, promotion, and further authorization remain unavailable.

[coderabbitai]

Warning

Review limit reached

@laiqian0239-glitch, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 51 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: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 6ad0f987-e37a-483e-b29b-c5add6a3bca6

📥 Commits

Reviewing files that changed from the base of the PR and between a45b11d and b2eb6c1.

📒 Files selected for processing (1)

• governance/layered-ci/v21-learning-policy-p1-decision-outcome-v3-production-closure-authorization.json

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: a469055c-f926-4d60-96d8-98be146a4713

📥 Commits

Reviewing files that changed from the base of the PR and between a45b11d and b2eb6c1.

📒 Files selected for processing (1)

• governance/layered-ci/v21-learning-policy-p1-decision-outcome-v3-production-closure-authorization.json

---

📝 Walkthrough

Walkthrough

The PR adds a V3 authorization document for the V21 Learning Policy P1 workflow. It defines trusted-main prerequisites, deterministic policy consumption, sealed runtime requirements, implementation scope, failure-first validation, prohibitions, and governance states.

Changes

V21 Learning Policy P1 authorization

Layer / File(s)	Summary
Authorization provenance and scope governance/layered-ci/v21-learning-policy-p1-decision-outcome-v3-production-closure-authorization.json	Defines V3 identity, trusted-main merge prerequisites, V2 supersession, prior failure evidence, fresh-main reconciliation, and single-file authorization scope.
Policy and sealed-runtime controls governance/layered-ci/v21-learning-policy-p1-decision-outcome-v3-production-closure-authorization.json	Defines adopted repository seams, deterministic policy selection, content-addressed artifacts, fallback behavior, and sealed Windows x64 runtime requirements.
Implementation scope and contracts governance/layered-ci/v21-learning-policy-p1-decision-outcome-v3-production-closure-authorization.json	Defines the implementation branch, allowed files, failure-first tests, dependency controls, required capabilities, and mandatory implementation contracts.
Prohibitions and governance states governance/layered-ci/v21-learning-policy-p1-decision-outcome-v3-production-closure-authorization.json	Prohibits bypasses, runtime downloads, request overrides, exploration, release, publishing, and automatic promotion. Records authorization and disabled production governance states.

Estimated code review effort: 4 (Complex) | ~45 minutes

Merge Risk: 🔵 Low · up to b2eb6

The change authorizes production closure without directly changing runtime behavior, but merge readiness still depends on confirming its declared content digests, validation contract, and pinned runtime assets. This is a bounded follow-up, so the PR is mergeable with explicit owner awareness.

Possibly related PRs

• laiqian0239-glitch/yance#368: Defines the V21 Learning Policy P1 direction and Decision→Outcome architecture referenced by this authorization.
• laiqian0239-glitch/yance#353: Defines related V21 Learning workflow closure controls, including trusted-main, deterministic, and failure-first requirements.
• laiqian0239-glitch/yance#347: Adds related V21 Learning successor authorization controls with strict implementation and prohibition rules.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely identifies the V21 Learned Policy V3 production-closure authorization added by the pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch governance/v21-learning-policy-p1-decision-outcome-v3-production-closure-authorization

Warning

Review ran into problems

🔥 Problems

Git: Failed to clone repository. Please run the @coderabbitai full review command to re-trigger a full review. If the issue persists, set path_filters to include or exclude specific files.

---

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.}

[laiqian0239-glitch]

@coderabbitai full review

[coderabbitai]

✅ Action performed

Full review finished.

[laiqian0239-glitch]

@coderabbitai full review

[laiqian0239-glitch]

@coderabbitai review

[coderabbitai]

⚠️ Action not completed

Already reviewed.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[coderabbitai]

⚠️ Action not completed

Review rate limited.

---

Your included review limit is currently reached under our Fair Usage Limits Policy. This review may still proceed through usage-based billing if eligible. Your next included review will be available in 51 minutes.