A set of project rules for Cursor, an AI coding editor. It tells the agent which project-status files to read and whether the work uses one or several workstreams, meaning separate areas of work.

Use it at the beginning of a coding session to inspect the project map and status, identify the active workstream, and report the current progress and blocker.

It prevents the agent from starting with the wrong project context or silently choosing the wrong workstream.

cursorrules — Cursor rule by Ethan-YS

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/ethan-ys/project-brain/cursorrules

Clone the repo

git clone --depth 1 https://github.com/Ethan-YS/project-brain

Made for: Cursor.

Per session 606 This file is loaded in full into every session.

When invoked 606 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	$0.00606	$0.00606
Opus 5	$0.00303	$0.00303
Sonnet 5	$0.00121	$0.00121
Haiku 4.5	$0.00061	$0.00061

Measured yesterday against content hash 68bc96891e40, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cursorrules 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 yesterday.

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/.cursorrules · 58 lines

How it starts

The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.

[Project Name] — Cursor Rules

This is the project-level instruction file for Cursor.

Layered above your global Cursor rules.

Write only what's specific to THIS project here.

First 30 seconds of a new session

Decide single- or multi-workstream (look at file naming under brain/):

Only STATUS.md / HANDOFF.md → single-workstream branch
Multiple STATUS_<workstream>.md → multi-workstream branch (v2.1)

Single-workstream (default)

Must read (under brain/):

brain/MAP.md — project map
brain/STATUS.md — current state
brain/HANDOFF.md if it exists — previous session's still-warm thoughts

Then briefly report: project recognized + current progress + last blocker. Don't silently start working — confirm understanding first.

Multi-workstream (v2.1)

First read shared files: brain/MAP.md + brain/PROJECT.md. Don't guess this window's workstream — ask the user: "I see this is a multi-workstream project with [list workstreams]. Which one does this window work on?" After the user says, read brain/STATUS_<workstream>.md + brain/HANDOFF_<workstream>.md.

If the user switches workstreams mid-session, re-read the new one's files and don't carry over memory.

Update responsibility

The AI doesn't silently modify any file in brain/ — propose, then user approves.

Judgment division:

The user decides "should we record now" (high-level pacing)
The AI decides "specifically what to record / how to write each entry"
The user has approval/rejection authority

Specific rhythm:

User says "update the project brain" → propose a list with reasons; user reviews
AI senses a decision was just made → gently inquire ("does this count as decided?"), don't assert
Module added/removed → propose updating brain/MAP.md §2
New doc added/removed → propose updating brain/MAP.md §5
User signals end-of-session → draft brain/STATUS.md for review (soft cap 80 lines)
User signals window-switch → write brain/HANDOFF.md, archive previous to brain/handoffs/YYYY-MM-DD-HHMM.md

Read the full file on GitHub · 58 lines

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.

yesterday First seen · 58 lines · 606 tokens per session scan A 68bc96891e40

Subscribe to this mod's changes

cursorrules is a cursor rule published in the GitHub repository Ethan-YS/project-brain (169 stars, last pushed 2mo ago), licensed MIT. It adds 606 tokens to every session, about $0.0030 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.

Memory and context ai-assistant ai-coding ai-pair-programming claude-code context-management cursor

cursorrules

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

[Project Name] — Cursor Rules

This is the project-level instruction file for Cursor.

Layered above your global Cursor rules.

Write only what's specific to THIS project here.

First 30 seconds of a new session

Single-workstream (default)

Multi-workstream (v2.1)

Update responsibility

What this file has done since we first saw it

Other cursor rules, from other repositories

cursorrules

cursorrules

workflows

review-checklist

guardrails

core-principles