What is cursor-brain-os?

A set of Cursor instructions for Brain OS, a system that stores project state and decisions in a .brain folder and exposes them through MCP tools.

It guides agents when reading project state, finding the next task, recalling decisions, detecting conflicts, and recording approved changes.

It prevents agents from guessing project priorities by searching source files and helps them follow shared instructions and decision checks.

cursor-brain-os — Cursor rule for coding agents by brainOS-HQ: A set of Cursor instructions for Brain OS, a…

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/brainos-hq/brain-os/cursor-brain-os

Clone the repo

git clone --depth 1 https://github.com/brainOS-HQ/brain-os

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/rules/brainos-hq/brain-os/cursor-brain-os.svg)](https://agentmods.dev/rules/brainos-hq/brain-os/cursor-brain-os)

Your own site

<a href="https://agentmods.dev/rules/brainos-hq/brain-os/cursor-brain-os"><img src="https://agentmods.dev/badge/rules/brainos-hq/brain-os/cursor-brain-os.svg" alt="Measured on agentmods" height="20"></a>

Per session 414 This file is loaded in full into every session.

When invoked 414 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.

Model	Per session	Once invoked
Fable 5	$0.00414	$0.00414
Opus 5	$0.00207	$0.00207
Sonnet 5	$0.00083	$0.00083
Haiku 4.5	$0.00041	$0.00041

Measured 4d ago against content hash 02724dd8b033, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cursor-brain-os 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 4d 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.

templates/agent-instructions/cursor-brain-os.mdc · 30 lines

What it actually says

Brain OS rules for Cursor

This project uses Brain OS for persistent operational state across AI agents. State lives in .brain/.

Canonical instructions

See AGENTS.md at the project root. It is the cross-tool source of truth. This .mdc file exists only so Cursor picks up the same rules.

Critical rules (also in AGENTS.md)

Brain OS MCP tools are the primary data source for any project-state question. The tools are exposed as brain-os in your MCP server list. Do NOT grep source code to answer "what should I focus on" — call mcp__brain-os__focus_get.
Routing order: entity_read → plan_read → focus_get → semantic_recall → decision_check → pattern_detect. Mutating tools (decision_log, entity_update, plan_*, decision_refresh) require a prior decision_check call.
Conflict handling: if decision_check returns conflict, do not proceed without explicit user confirmation. If caution, surface the relevant active decision and ask.
Output format: fixed-column tables for focus_get, single-card format for entity_read. No long-form rephrasing of JSON.
Command vocabulary: the user may type bare keywords (brain, focus, decide, wrap, retro, patterns, graph, strategy) with or without a leading /. Run the matching MCP tool sequence — do not search the repo for these words.
No todo lists for state-reading questions. Call the tool, return the answer.
Drift check: if your first action on a state question is reading .brain/ files directly, stop and re-route through the MCP tools.

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.

4d ago First seen · 30 lines · 414 tokens per session scan A 02724dd8b033

Subscribe to this mod's changes

cursor-brain-os is a cursor rule published in the GitHub repository brainOS-HQ/brain-os (6 stars, last pushed 11d ago), licensed MIT. It adds 414 tokens to every session, about $0.0021 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 2026-08-31.

Memory and context agent-memory ai-agents claude-code cursor developer-tools local-first

cursorrules

Zettelkasten memory for AI agents. Markdown cards in /.memex/cards/ with [[wikilinks]]. npm package @touchskyer/memex. Distributed as CLI + MCP server + Claude Code plugin + VS Code extension + Pi extension.

iamtouchskyer/memex · 1,610 tokens

lians-memory

Current Lians task, control policy, and bounded context.

Lians-ai/Lians · 223 tokens

ctxfile

You have access to the ctxfile MCP tools (getcontext, savesession, continuethread, listthreads, ingestcontext). These behaviors make context capture automatic. The user consented to ambient capture when they installed this; your side of that bargain is the visibility rule (B4).

ctxfile/ctxfile · 983 tokens

cursor-brain-os