Repository navigation
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Code coverage overviewLanguages: TypeScript TypeScript / code-coverage-agent
View coverage of the 10 most impacted files.
ℹ️ Calculations represent line coverage. Results may change as calculations are updated. |
This branch was successfully deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The Local Models implementation and native-recovery stacks were preserved on separate branches while main gained later conversation-fidelity repairs. This integrates their complete implementation history on current main, keeps production local execution unadmitted, and adds secure packaged cipher selection plus an accessible follow-up proposal label.
The PR also adds process-memory observations to the existing transcript E2E probe, corrects an outdated Local Models guide paragraph, and restores the missing 33-image October 22 CF-8 manifest target byte-for-byte under the original owner's scoped handoff. The historical captures retain their original candidate and fixture provenance. Current qualification and its limits retain each evidence class and original failure.
Validation of executable candidate
7d6094e: build/typecheck passed; 2,407 units passed with 35 skips; 508 desktop E2E passed with 11 opt-in skips and seven site cases passed; four focused Local Models journeys passed. Separate opt-ins passed three Open Knowledge Electron cases, one pinned-runtime unit and six native Code Review confinement cases. Codex 0.167.0/Ollama 0.40.16 passed the network-denied real-model read/edit/check canary and all four real-model Electron lifecycle cases, with unchanged personal configuration and 65,536 loaded context. The new canonical memory spec passed two cases; its cold median was 76.4 ms against the unchanged 500 ms gate. The source audit found no executable-input changes in the test/documentation increment, and the 213-file Markdown link audit found no missing relative targets or unmatched anchor candidates.First failures remain retained: the initial offline workspace was refused by the existing ancestor
.codexguard and only that gate was retried in private system temp; the first memory collection accidentally included the nested historical baseline. The baseline correctly fails the newer output bound. The packaged credential check remains unaccepted because main waited on native Keychain access; no OS prompt was bypassed or approved. Memory observations show bounded IPC output but do not establish lower memory use or a strict ceiling.No issue closure or product activation is claimed. The LM product decisions, other-runtime/compaction/OOM/managed-policy qualification, human and physical-device acceptance, Code Review live native/write gates, hosted checks and protected delivery remain separate. PR #475 can be reconciled only after replacement scope is on main. PRs #497/#496 delivery is independent and is not repaired by this integration. No release, tag, TestFlight upload, installed-app replacement or other-owner cleanup is included.