Documentation Index

All Bernstein documentation in one place. Guides, references, design decisions, deployment, and comparisons.

Guides & Setup

DocumentDescription
Getting StartedInstall, quick start, configuration, CLI reference, task API
The Bernstein WayArchitecture tenets and default workflow philosophy
Linux Install GuidePlatform-specific installation notes for Linux
GitHub App SetupStep-by-step GitHub App installation and webhook configuration
GitHub ActionCI-triggered orchestration via GitHub Actions

Architecture & Design

DocumentDescription
ConceptsTask server, spawner, janitor, agent lifecycle, model routing
Design DocumentFull system architecture, component interactions, data flow
Dependency ResolverHow task dependency graphs are resolved and scheduled
RAG Challenge SwarmOrigin story: 12 agents, 47 hours, 737 tickets, architectural lessons

Deployment

DocumentDescription
Deployment GuideProduction deployment patterns, storage backends, environment config
Cluster ModeDistributed multi-node orchestration with worker daemons
Helm DeploymentKubernetes deployment via Helm charts

Protocols & Compatibility

DocumentDescription
Protocol Compatibility MatrixMCP 1.0/1.1, A2A 0.2/0.3, ACP — test results and version support
Zero Lock-InHow Bernstein avoids provider and protocol lock-in

Governance & Policy

DocumentDescription
Model Routing PolicyProvider allow/deny lists, model selection rules, cost controls
Known LimitationsCurrent constraints, workarounds, and planned fixes

Extensibility

DocumentDescription
AdaptersAgent adapter reference: all supported CLIs, feature matrix, custom adapters
Plugin SDKPluggy-based plugin system: hooks, lifecycle events, packaging
API ReferenceTask server HTTP API: endpoints, schemas, config

Comparisons

DocumentDescription
Comparison IndexOverview of all comparison pages with selection guidance
vs. ConductorWorkflow engine comparison
vs. CrystalMulti-agent framework comparison
vs. DorothyOrchestration system comparison
vs. GitHub Agent HQGitHub's multi-agent system comparison
vs. Parallel CodeParallel coding tool comparison
vs. Single AgentMulti-agent orchestration vs. single-agent baseline
vs. StoneforgeAgent platform comparison
Competitive MatrixFeature-by-feature comparison table across tools

Coverage & Status

DocumentDescription
Feature MatrixDocumentation coverage status for every shipped capability
ChangelogRelease notes and version history

Design Decisions

DocumentDescription
001 — Agent LifecycleWhy short-lived agents over long-running processes
002 — UX and DistributionCLI-first design, distribution strategy
003 — Self-EvolutionRisk-gated self-improvement architecture

Blog & Write-ups

DocumentDescription
Zero-LLM CoordinationHow Bernstein achieves multi-agent orchestration without LLM scheduling
Multi-Agent BenchmarkBenchmark results: orchestrated vs. single-agent performance
Self-Evolution: 30 DaysResults from 30 days of continuous self-improvement
SWE-Bench OrchestrationApplying orchestration patterns to SWE-bench tasks