This release adds 5 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+6 more
Summary
AI summaryInitial public release of Galley, a local orchestration runtime for supervised Claude Code task execution.
Full changelog
Galley 🚢
Initial public release of Galley, a local orchestration runtime for supervised Claude Code task execution.
Galley runs repository tasks from YAML files, executes them in isolated git worktrees, records run evidence, and asks a model supervisor to accept, retry, or escalate the result before completion.
Highlights
- File-backed task queue under
~/.galley - Background daemon with graceful shutdown, heartbeat, stale-claim handling, and retry loops
- Claude Code executor with structured result validation
- Claude and Codex supervisor adapters
- Quality and environment profiles for repo-specific checks and execution constraints
- Run evidence under
runs/, including command plans, logs, diffs, profile snapshots, and supervisor verdicts - Optional commit, PR creation, PR comment polling, and conservative worktree cleanup
- Galley plugin and Agent Skill for task authoring, profile authoring, setup, and troubleshooting
- Release binary installer plus
go installsupport
Install
curl -fsSL https://raw.githubusercontent.com/shinpr/galley/main/scripts/install.sh | sh
Go users can also install with:
go install github.com/shinpr/galley/cmd/galley@latest
Notes
This is an early preview. Galley is intended for trusted local repositories and trusted task authors. Task YAML and profiles are privileged local execution inputs.
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]