Repository navigation
Release the adaptive review controller with operating guidance - #7
Merged
Merged
Conversation
Assign focused native reviewers to pinned evidence, validate hunk coverage and input budgets, and gate optional improvements on verified essentials. Keep the initial 58,000-byte capacity explicitly provisional. Provide isolated model onboarding, retained trial evidence and measured profile promotion without running paid calibration campaigns. Validate scheduling and onboarding with synthetic workers and 54 offline tests.
Replace scaffold and feature-branch instructions with the first tagged release installation target. State the supported Linux and Python range, private artifact requirements, gate meanings, and interruption recovery. Keep provisional capacity limits and unavailable measurements explicit.
Include the newly stable interpreter in release support guidance to match the validated Python 3.11 through 3.15 CI matrix.
Run the existing package checks and offline tests on Python 3.11 through 3.15 so the release has the validation documented in its operating guide. Reuse the established workflow and pinned actions from the project.
Keep Docker stdin attached so the review prompt reaches the trial. Record interrupted attempts before propagating SIGINT through existing cleanup and receipt persistence.
Prevent Git settings, optional prompt overhead, and incomplete campaign measurements from producing invalid coverage or profiles. Validate fixture boundaries and keep standalone skill state outside the actual package.
Calibration must retain raw pinned contents and executable modes. Keep reserved time charged after interrupted trials, remove identifiable version probes on every exit, and keep promoted profiles outside each original fixture repository.
Check node types before hashing so a substituted symlink cannot keep a campaign identity while pointing outside its mounted source. Preserve valid file contents and executable masks.
Keep frozen base/head trees complete for nested repository paths and empty revisions. Retain incompatible native campaigns as historical leads without weakening exact-identity reuse, and exercise prompt drift apart from answer-key changes.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Issue
The public main branch contains only a scaffold, and installation instructions point users at an implementation branch without operating or recovery guidance.
Fix
Integrate the standalone adaptive review skill and its deterministic controller, onboarding harness, and focused guidance. Document the v0.1.0 stable installation target, supported environment, private state, gate meanings, and interruption recovery while retaining provisional capacity claims.