BE
Deterministic orchestration for parallel CLI coding agents
Bernstein runs multiple CLI coding agents in isolated git worktrees, then verifies and merges only the diffs that pass checks. It is designed for reproducible, on-prem agent workflows with signed audit logs and offline review.
Agenticness = how independently a tool can take action, scored across 9 dimensions. Scored independently by David Kooi, Skylark Creations — see full rubric → Last re-evaluated 12 August 2026.
Open Source
Chrome Extension
Multi-Agent
B2B
CLI
Self-Hosted
On-Device / Edge
Is this your tool? Claim this listing to manage your content and analytics.
About
What It Is
What to Know
Key Features
Runs Claude Code, Codex, Gemini CLI, Aider, and 40+ CLI agents
Spawns each agent in its own isolated git worktree
Decomposes goals into tasks and schedules them in parallel
Gates merges on lint, type checks, and tests
Supports optional cross-model review before merge
Use Cases
Coordinating several coding agents on the same repository without shared-state conflicts
Running test-gated code changes across backend, QA, security, or devops tasks
Reviewing agent output offline using signed audit logs
Agenticness
: Domain Specialist
Handles domain-specific workflows independently with dynamic replanning.
High evidence
Last evaluated:
Aug 12, 2026
Dimension Breakdown
Action Capability
Autonomy
Planning
Adaptation
State & Memory
Reliability
Interoperability
Safety
Operator Sovereignty
Categories
Pricing
- Free / open source — full functionality available at no cost.
Details
AddedAugust 12, 2026
RefreshedAugust 25, 2026
Agenticness
Quick Facts
DeploymentHybrid (cloud + self-hosted)
AutonomySemi-autonomous
Model supportMulti-model
Open sourceYes
Team supportEnterprise
Pricing modelFree / open source
Interfacecli
Sources
Last updated
August 26, 2026
Free daily brief
When Bernstein changes its pricing, we catch it the next morning
We recrawl every tool in this directory daily. AI Insider is the 3-minute brief on what actually changed — pricing, features, score moves on our 36-point rubric — often before any press coverage. Sent daily. Skipped when nothing happened.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.