This release adds 1 notable feature for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+6 more
Summary
AI summarySupervisor reviews now require evidence for acceptance-relevant boundary paths when behavior-changing criteria could diverge.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Medium |
Supervisor reviews now require evidence for acceptance-relevant boundary paths when a behavior-changing criterion could diverge. Supervisor reviews now require evidence for acceptance-relevant boundary paths when a behavior-changing criterion could diverge. Source: llm_adapter@2026-06-03 Confidence: high |
— |
Full changelog
Patch release focused on stronger supervisor evidence for behavior-changing work.
What's Changed
- Supervisor reviews now require evidence for acceptance-relevant boundary paths when a behavior-changing acceptance criterion could pass on the main path while failing a separate contract dimension.
- This makes accepted task reviews less likely to miss edge-specific regressions that are not covered by the happy-path implementation or verification evidence.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Galley
Local-first runtime for supervised AI coding tasks: isolated git worktrees, supervisor review against acceptance criteria, retry/escalate loops, on-disk run evidence, and PR handoff. Supports Codex CLI and Claude Code. Go, MIT.
Related context
Beta — feedback welcome: [email protected]