bernstein
The open‑source AI Agents Governance & Orchestration framework: write the rules declaratively, Bernstein enforces them and produces the verifiable, replayable record. Free, Apache-2.0. https://bernstein.run
- stars
- 1,186
- 30-day movement
- +4515/day
- Related entries
- 62
- Connections
- 8
Bernstein is an open-source AI agent governance and orchestration framework that you configure by writing rules declaratively; it enforces those rules and produces a verifiable, replayable record of agent activity. It ships as an MCP server with a Python toolchain (Docker, uv) and includes config artifacts such as agents.md, claude.md, cursor rules, devcontainer, and MCP config/manifest.
Reach for it when you need declarative rule enforcement over AI agents with a tamper-evident, auditable trail for compliance or reproducibility.
Use it to
- Enforce declarative governance rules on agent fleets
- Produce tamper-evident audit logs and attestations
- Run deterministic replay of agent sessions
- Orchestrate parallel agents using git worktrees
- Operate agents in air-gapped environments
For Teams running AI agents under compliance or audit requirements
- Role
- mcp-server
- Language
- Python
- Licence
- Apache-2.0
- Forks
- 161
- Open issues
- 251
- Last push
- 2026-09-17
- Latest release
- v1.0.0 · 2026-03-29