12 added, 9 removed. Audit A to A.
# Bernstein
- > Forward-deployed engineering, with a coding swarm that fits in your `.sdd/`.
+ > Bernstein is a deterministic orchestrator for CLI coding agents (Claude Code, Codex,
+ > Gemini CLI, and 40+ more). Scheduling is plain Python - no LLM in the coordination
+ > loop - so runs are reproducible end to end.
>
- > Bernstein is an open-source multi-agent orchestration platform for AI coding agents.
- > Deterministic scheduling, 40+ CLI adapters, pluggable sandbox backends, cloud artifact
- > storage sinks, progressive-disclosure skill packs, zero vendor lock-in. Apache 2.0 licensed.
+ > Every task runs in its own git worktree behind lint/type/test gates. An always-on
+ > lineage spine and replay journal record what happened; an opt-in HMAC-chained audit
+ > log and signed receipts make results independently verifiable, offline if needed.
+ > Air-gap install profile included. Apache-2.0.
- ## Forward-deployed engineering use case
+ ## How it runs
- Forward-deployed engineering, on a swarm. Drop Bernstein into a
- client repo and you get a multi-agent crew with file-based state,
- per-agent credential scoping, and an HMAC-signed audit trail -
- running on whichever CLI agents the client already trusts.
+ Drop Bernstein into a repo and you get a multi-agent crew with file-based
+ state in `.sdd/`, per-agent credential scoping, and per-artefact lineage -
+ running on whichever CLI agents you already trust. Enable `--audit` for an
+ HMAC-signed audit trail on top.
## Core Documentation
- [Getting Started](https://bernstein.readthedocs.io/getting-started/install/): Install, configure, run first orchestration in 60 seconds
- [Architecture & Concepts](https://bernstein.readthedocs.io/architecture/ARCHITECTURE/): Task server, agent spawner, janitor verification, worktree isolation
- [Adapter Reference](https://bernstein.readthedocs.io/adapters/): 40+ supported agents - Claude Code, Codex, OpenAI Agents SDK v2, Gemini CLI, Cursor, Aider, Devin Terminal, CLM gateway, and more
- [API Reference](https://bernstein.readthedocs.io/reference/openapi-reference/): 120+ REST endpoints for tasks, agents, cost tracking, quality gates
## Guides
- [Benchmarks](https://bernstein.readthedocs.io/benchmarks/BENCHMARKS/): SWE-Bench Lite results for solo agents and orchestrated multi-agent runs
- [Known Limitations](https://bernstein.readthedocs.io/reference/KNOWN_LIMITATIONS/): Honest constraints and workarounds
## Resources
- [GitHub Repository](https://github.com/sipyourdrink-ltd/bernstein): Source code, issues, discussions
- [PyPI Package](https://pypi.org/project/bernstein/): pip install bernstein
- [Blog](https://bernstein.readthedocs.io/blog/ten-agents-one-release/): Engineering articles on multi-agent orchestration patterns
- [Author](https://alexchernysh.com): Built by Alex Chernysh ([@alex_chernysh on X](https://x.com/alex_chernysh))