architecture
01Cursor rule Cursor
Top-level entry points exposed by the package.
Cursor rule Cursor
Top-level entry points exposed by the package.
Cursor rule Cursor needs its repo
Build & test.
Cursor rule Cursor
Per-directory context files for agents working inside a subtree. Each covers: what the directory is for, key files, invariants, and how to test that subtree.
Cursor rule Cursor
Every PR that adds or changes a feature MUST update docs in the same PR.
Cursor rule Cursor
Git workflow.
Cursor rule Cursor
Full per-file map: docs/sdd/module-map.md (generated).
Cursor rule Cursor
Bernstein is the open-source governance layer for AI agents. It runs on policy as code: you write the policy - who may do what, what needs approval, what must be recorded - and Bernstein enforces it and produces the verifiable record. A deterministic scheduler - no model in the coordination loop - runs agents in…
Cursor rule Cursor
Bernstein ships agent role prompts under templates/roles/. The orchestrator loads them at task-spawn time; you don't write to them manually.
Cursor rule Cursor
Setup.
MCP server Claude CodeCodexCursor +2
MCP server "bernstein" as configured in sipyourdrink-ltd/bernstein. Launched with bernstein mcp.
Agent Claude Code
Decomposes goals into parallel tasks, assigns them to CLI coding agents, verifies output, and merges results. Use when a task is too large for a single agent.
Command Claude Code
Start a Bernstein orchestration run with a goal.
Command Claude Code
Show current Bernstein orchestration status.
Command Claude Code
Gracefully stop a running Bernstein orchestration.
Cursor rule Cursor
Bernstein governance-layer context - always active when working in a Bernstein-managed project.
Cursor rule Cursor
Guidelines for working in a Bernstein-orchestrated project.
Skill Claude CodeCodex
Run a verified multi-agent goal with Bernstein. Use when a task is too large for a single agent session: Bernstein decomposes the goal into tasks, spawns CLI coding agents in parallel git worktrees, verifies their output, and merges results. Also use to check run status, costs, and to verify a finished run against its…
Skill Claude CodeCodex
Evaluate proposals - feasibility, engineering payoff, risk.
Skill Claude CodeCodex
System design - module boundaries, API contracts, ADRs.
Skill Claude CodeCodex
Python server code, APIs, async, strict typing.
Skill Claude CodeCodex
CI failures - read error, minimal fix, verify.
Skill Claude CodeCodex
DevOps - Docker, CI/CD, cloud infra, monitoring.
Skill Claude CodeCodex
Documentation - README, API docs, ADRs, tutorials.
Skill Claude CodeCodex
React / Next.js UI, state, accessibility.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: