bernstein-context

Background rules for Bernstein, a tool that coordinates coding agents in separate workspaces and tracks their tasks, checks, and costs. It explains the task states and where orchestration data is stored.

In plain words
What is it for?
Checking task progress, creating or approving coordinated work, reviewing costs, and following required lint, type-check, and test gates.
Why use it?
They help an agent understand how Bernstein-managed work is organised and which status or task workflows to use.

Cursor rule

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add rules/sipyourdrink-ltd/bernstein/bernstein-context
Clone the repo
git clone --depth 1 https://github.com/sipyourdrink-ltd/bernstein
Per session 399 This file is loaded in full into every session.
When invoked 399 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00399 $0.00399
Opus 5 $0.00199 $0.00199
Sonnet 5 $0.00080 $0.00080
Haiku 4.5 $0.00040 $0.00040

Measured 2d ago against content hash 589684571676, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

bernstein-context scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

packages/cursor-plugin/rules/bernstein-context.mdc · 37 lines

What it actually says

Bernstein Orchestrator Context

You are working in a project managed by Bernstein, a deterministic orchestrator for CLI coding agents. Bernstein runs at http://127.0.0.1:8052 and coordinates parallel CLI coding agents, one git worktree per task. There is no model in its coordination loop, so the same plan replays to a byte-identical task graph.

Key concepts

  • Tasks have states: openclaimedin_progressdone | failed | cancelled | blocked
  • Agents are short-lived (1-3 tasks each), spawned per task, isolated in git worktrees
  • Roles: backend, frontend, qa, security, devops, docs - each agent specializes
  • Quality gates: every task goes through lint, type-check, and test verification before completion
  • Cost tracking: real-time per-model, per-agent spend with budget enforcement

State lives in files

  • .sdd/ directory contains all orchestrator state (backlog, runtime, metrics, config)
  • .sdd/backlog/ has task YAML files
  • Agents are spawned fresh per task - no persistent memory

Use the /bernstein-status, /bernstein-create-task, /bernstein-approve, or /bernstein-cost skills to interact with the orchestrator API at http://127.0.0.1:8052.

API base URL

The Bernstein API runs at http://127.0.0.1:8052 by default. Set the BERNSTEIN_API_URL environment variable to override. Set BERNSTEIN_API_TOKEN for authenticated setups.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 37 lines · 399 tokens per session scan A 589684571676

Subscribe to this mod's changes

bernstein-context is a cursor rule published in the GitHub repository sipyourdrink-ltd/bernstein (1,038 stars, last pushed yesterday), licensed Apache-2.0. It adds 399 tokens to every session, about $0.0020 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.