Bernstein — Multi-Agent Orchestration

Version updated for https://github.com/sipyourdrink-ltd/bernstein to version v3.17.2.

  • This action is used across all versions by 5 repositories.

Action Type

This is a Composite action.

Go to the GitHub Marketplace to find the latest changes.

Action Summary

Bernstein is a deterministic multi-agent CLI orchestration tool that automates the deployment and management of complex systems by orchestrating multiple agents to execute tasks in parallel. It solves the problem of managing distributed applications with many interconnected components, ensuring consistency and reliability. The key capabilities include defining workflows, managing dependencies between tasks, and monitoring agent performance.

What’s Changed

A patch release about proof: what a run recorded, and whether you can still check it afterwards.

Receipts you can re-derive

A gate verdict is three-valued, but receipts stored only the outcome, so an auditor re-running the check offline could not tell a refusal from an abstention. Receipts carry the verdict now, and it re-derives without the run (#4182).

Spawn context was assembled from a dozen sources and hashed as one blob. Each part now gets its own content hash in a receipt, in prompt order, so a prompt that changed can be narrowed to the section that changed it (#4303). An agent working inside a subtree is handed that subtree’s .sdd/project.md rather than only the root’s (#4304). The breakdown surface that never got populated is gone (#4422).

A persistent-agent adapter carries state Bernstein never hashed, so replaying its inputs proves nothing. Those runs mark their artifacts unverifiable rather than verified (#4290), and letta_code declares itself as one (#4289).

The receipt ledger has a pure active-set closure (#4183). LineageGate.check verifies who wrote a permission-bearing file, not what the write grants — the two are different questions and it was answering the wrong one (#3768). Admission receipts were never written at all for an adapter whose name contains a space; the filename is slugged now (#4363). A CAS sidecar read skipped the anchored walk on an assumption nothing enforced (#3582).

Runs that lied about themselves

A planning task that decomposed into nothing reported the run as successful. It fails now (#4401). Three /status surfaces counted live agents three different ways, so the same run read as busy on one and idle on another (#4360). A task stranded by a failed dependency stayed stranded after that dependency was retried and succeeded (#4376).

A skip-worktree CLAUDE.md left by a killed run survived reset --hard and poisoned every later run in that checkout (#4394).

Models and transport

Every 429 was backed off as a rate limit, including standing account caps that no retry can clear — the agent spent its budget waiting for a wall to move (#4378). The last-green adapter table showed a never-probed adapter as merely stale (#4387).

The assembled system prompt now has a spawn-time budget: over a configurable share of the model’s context window it warns, naming the sections responsible (#4377).

Security

A resume whose grant moved is refused before the first side effect, not after (#3834). The skills and MCP catalog fetchers reject internal-host destinations (#4301, #4302).

Also

RPM never published when the install smoke outran PyPI index propagation (#4383). The metric buffer dropped every target’s lines when one target’s write failed (#3710). Coverage baseline is 84.08%. Agent working notes under scratch/ no longer ride along in unrelated commits.

Soundtrack: https://suno.com/s/mlHRUsFOZfhEUL96

What’s Changed

New Contributors

Full Changelog: https://github.com/sipyourdrink-ltd/bernstein/compare/v3.17.1...v3.17.2