cursor

cursor is a cursor rule for coding agents from proticom/gnosys. It costs 990 tokens per session, scanned A, original, MIT.

A set of project rules for using Gnosys, a shared persistent memory system for coding agents. It explains how to search, read, and save project, user, and global memories.

In plain words
What is it for?
Use it to discover relevant past notes, load decisions, search across projects, inspect recently changed memories, and save structured information for future tasks.
Why use it?
It helps an agent retain useful decisions and retrieve earlier context instead of treating every task as new. It also defines which memory operations to use and how to structure saved entries.

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/proticom/gnosys/cursor
Clone the repo
git clone --depth 1 https://github.com/proticom/gnosys

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.

agentmods badge for cursor

README.md
[![agentmods](https://agentmods.dev/badge/rules/proticom/gnosys/cursor.svg)](https://agentmods.dev/rules/proticom/gnosys/cursor)
Your own site
<a href="https://agentmods.dev/rules/proticom/gnosys/cursor"><img src="https://agentmods.dev/badge/rules/proticom/gnosys/cursor.svg" alt="Measured on agentmods" height="20"></a>
Per session 990 This file is loaded in full into every session.
When invoked 990 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 $0.00990 $0.00990
Opus 5 $0.00495 $0.00495
Sonnet 5 $0.00198 $0.00198
Haiku 4.5 $0.00099 $0.00099

Measured yesterday against content hash a20e3c85b0f8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cursor 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.

src/test/fixtures/ide-init/cursor.mdc · 52 lines

How it starts

The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Gnosys Memory System

This project uses Gnosys for persistent memory via MCP. Gnosys uses a centralized brain (~/.gnosys/gnosys.db) shared across all projects with project, user, and global scopes.

Read first

  • At task start, call gnosys_discover with relevant keywords
  • Load results with gnosys_read
  • When the user references past decisions, says "recall", "remember when", "what did we decide" — search memory first
  • Use gnosys_federated_search for cross-project search with scope boosting
  • Use gnosys_working_set to see recently modified memories for context
  • Gnosys starts on a reduced core toolset — if a gnosys tool you need is missing, call gnosys_toolset (set: standard or full) to expand it

Write automatically

  • Always write with gnosys_add_structured — you are an LLM, so structure the memory yourself (title, category, tags, content, relevance) and pass the explicit fields. NEVER call the freeform gnosys_add: the server rejects it with an error for MCP callers (it would run a second, redundant LLM call and depends on a provider key that can fail silently). gnosys_add_structured makes no server-side model call. (Freeform gnosys_add exists only for non-agent scripts/cron, which must set GNOSYS_ALLOW_FREEFORM_ADD=1.)
  • When user says "remember", "memorize", "save this", "note this down", "don't forget" — call gnosys_add_structured
  • When user states a decision or preference (even casually) — commit to decisions category
  • When user provides a spec or plan — commit BEFORE starting work
  • After significant implementation — commit findings and gotchas
  • User preferences (coding style, conventions) — use gnosys_preference_set

Key tools

Action Tool
Find memories gnosys_discover (metadata) → gnosys_read (content)
Search gnosys_hybrid_search (best), gnosys_federated_search (cross-project), gnosys_search (keyword), gnosys_ask (Q&A)
Write gnosys_add_structured (explicit fields — always use this as an agent; no server-side LLM call). gnosys_add (freeform) is rejected for MCP callers — non-agent scripts only.
Update gnosys_update, gnosys_reinforce (useful/not_relevant/outdated)
Browse gnosys_list, gnosys_lens (filtered), gnosys_tags, gnosys_graph
Maintain gnosys_maintain, gnosys_stale, gnosys_history, gnosys_dashboard
Preferences gnosys_preference_set, gnosys_preference_get, gnosys_preference_delete
Projects gnosys_init (register), gnosys_briefing (status), gnosys_stores (debug)
Context gnosys_federated_search, gnosys_working_set, gnosys_detect_ambiguity
Recall gnosys_recall (fast context injection, sub-50ms)
Export gnosys_export (Obsidian vault), gnosys_audit (operation trail)

Read the full file on GitHub · 52 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.

  1. yesterday First seen · 52 lines · 990 tokens per session scan A a20e3c85b0f8

Subscribe to this mod's changes

cursor is a cursor rule published in the GitHub repository proticom/gnosys (4 stars, last pushed yesterday), licensed MIT. It adds 990 tokens to every session, about $0.0049 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 2026-09-03.