Repository navigation
Fix Windows startup, sandbox command completion, and editor regressions - #11
Merged
Merged
Conversation
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars.
|
ShadowfetchStudio
marked this pull request as ready for review
October 31, 2026 16:37
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Windows 2.3.0 failed to start on a PC without the Microsoft CRT, and the managed model runtime depended on unbundled OpenSSL DLLs. After startup was repaired, sandboxed Python commands could report success with empty output because BusyBox exited before the child finished. This change bundles the required CRT, disables unused runtime HTTPS support, and waits for real command output and exit status inside AppContainer.
It also fixes Windows worktree shell selection, Ctrl+Shift+Z redo while retaining Ctrl+Y, debug desktop stack exhaustion, and fresh-process GGUF selection. Removed configured model files cannot be selected through a stale cache. Windows test fixtures now handle clipboard/Git line endings, time zones and overlapping fake-vendor stderr logs. The compact exec schema retains cwd guidance while fitting the original 4,096-token tester limits.
Linux CI retains all checks while reducing debug-object disk usage. The AppImage build container's unavailable Alpine package revisions are refreshed against both architecture indexes. Retained musl/zlib source records match their linked package revisions, including all distribution patches; 19 source files were checked against APKBUILD SHA-512 values. The source-fetch regression passes. Main's documentation updates are merged, including published-version download instructions and support links.
Validation:
Remaining release gates: current-head CI, final release publication workflow and qualification of the exact public Windows installer (or proof that its bytes match a qualified artifact). Real cloud-provider accounts, Windows ARM64 and Windows 10 were not exercised. Windows GPU acceleration is not qualified. Website files are staged with a public-asset checksum gate; the site owner will publish them, and this task leaves the live website unchanged.