Cette page est affichée en anglais. Une traduction française est en cours.
WorkflowsMay 13, 2026·2 min de lecture

Bernstein — Audit-Grade Orchestrator for CLI Agents

Bernstein coordinates CLI coding agents in parallel worktrees with signed audit chains, deterministic scheduling, and evidence trails.

Agent Toolkit
Agent Toolkit · Community
Prêt pour agents

Cet actif peut être lu et installé directement par les agents

TokRepo expose une commande CLI universelle, un contrat d'installation, le metadata JSON, un plan selon l'adaptateur et le contenu raw pour aider les agents à juger l'adaptation, le risque et les prochaines actions.

Native · 94/100Policy : autoriser
Surface agent
Tout agent MCP/CLI
Type
Workflow
Installation
Pipx
Confiance
Confiance : Established
Point d'entrée
pipx install bernstein
Commande CLI universelle
npx tokrepo install 4c5d8b93-08ae-5a4b-bda3-2d5d5db42d12
contrat d'installationJSON metadataplan adaptateurcontenu raw
Introduction

Bernstein coordinates CLI coding agents in parallel worktrees with signed audit chains, deterministic scheduling, and evidence trails.

Best for: teams that need parallel agent execution with strong lineage, compliance evidence, and reproducible scheduling rather than informal orchestration

Works with: pipx, Git worktrees, Claude Code, Codex, Gemini CLI, multi-agent scheduling, enterprise audit requirements

Setup time: 10-18 minutes

Key facts (verified)

  • GitHub: 334 stars · 40 forks · pushed 2026-05-12.
  • License: Apache-2.0; owner avatar verified from GitHub API for sipyourdrink-ltd.
  • Entry point checked from README: pipx install bernstein.

Main

Bernstein is aimed at organizations that cannot treat orchestration as a black box. Every scheduled action is meant to leave evidence strong enough for security or compliance review.

The headline numbers are practical signals: 43 agent adapters, HMAC-SHA256 audit chaining, and deterministic worktree scheduling make it more than a wrapper around shell scripts.

Adopt it when several coding agents are already in play and the next problem is not access to models but control, lineage, and post-incident reconstruction.

Source-backed notes

  • README calls Bernstein a deterministic Python scheduler for CLI coding agents.
  • The project documents 43 CLI agent adapters in v1.10.1.
  • The security model highlights an HMAC-SHA256 audit chain and signed agent cards.

FAQ

Q: Is Bernstein only for big enterprises? A: No, but it is most valuable when teams care about auditable orchestration and reproducible evidence.

Q: How fast is first use? A: The README shows a 30-second install path via pipx install bernstein, then init and run.

Q: What is the main differentiator? A: Deterministic scheduling plus signed audit trails across many CLI agents.

🙏

Source et remerciements

Source: https://github.com/sipyourdrink-ltd/bernstein > License: Apache-2.0 > GitHub stars: 334 · forks: 40

Fil de discussion

Connectez-vous pour rejoindre la discussion.
Aucun commentaire pour l'instant. Soyez le premier à partager votre avis.

Actifs similaires

Claudable — Web Builder Powered by Local CLI Agents

Claudable is an open-source web builder that leverages local CLI agents (Claude Code, Codex, Gemini) with a repo-first quick start (3,951★).

Workflows
AI Open Source

RubyLLM::Agents — Rails Engine for Production Agents

Rails engine for building and monitoring LLM agents with a DSL, retries/fallbacks, cost analytics, and a dashboard; verified 124★, pushed 2026-04-19.

Workflows
Agent Toolkit

ZenGuard — Runtime Guardrails for AI Agents

A real-time trust layer for agents with prompt-injection/PII/secrets detectors and tiered access; verified 150★, pushed 2026-02-03.

Workflows
Agent Toolkit

agent-audit — Security Linter for LLM Agents

Run a static security scanner for LLM agents: 53 OWASP Agentic Top 10 rules, prompt-injection checks, and MCP config auditing via agent-audit scan.

CLI Tools
Agent Toolkit
◈Accueil🔍Rechercher👤Moi