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.
git clone --depth 1 https://github.com/mehmetdemirci/comb-aiWrote 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.
[](https://agentmods.dev/rules/mehmetdemirci/comb-ai/create-memory-bank)<a href="https://agentmods.dev/rules/mehmetdemirci/comb-ai/create-memory-bank"><img src="https://agentmods.dev/badge/rules/mehmetdemirci/comb-ai/create-memory-bank/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/mehmetdemirci/comb-ai/create-memory-bank"><img src="https://agentmods.dev/badge/rules/mehmetdemirci/comb-ai/create-memory-bank.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.00036 | $0.00694 |
| Opus 5 | $0.00018 | $0.00347 |
| Sonnet 5 | $0.00007 | $0.00139 |
| Haiku 4.5 | $0.00004 | $0.00069 |
Grade A, and why
create-memory-bank 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create Memory Bank — Cursor Rule (Greenfield)
Trigger: User says "create memory bank", "initialize memory bank", or "setup memory bank". This rule is NOT always-apply — it activates only when Cursor detects a matching request.
For existing projects with code: Use "adopt memory bank" instead — it analyzes the codebase and pre-fills memory bank files with real data.
When Triggered
Step 1 — Gather project information
Ask the user:
- Project name: What is the project called?
- Technical domains: What are the main development areas? (e.g., backend, frontend, infrastructure)
- Tech stack per domain: What technologies are used?
- Current phase: Kickoff, active development, or maintenance?
Step 2 — Create directory structure
memory-bank/
├── INDEX.md
├── core/
│ ├── projectbrief.md
│ ├── productContext.md
│ ├── active-context.md
│ └── sprint-tracker.md
├── tech/
│ └── (one file per domain)
└── arch/
├── patterns-overview.md
└── decisions-summary.md
Step 3 — Create project root files
memory-bank-protocol.md— COMB methodology (readmemory-bank-protocol.mdfor template)project-rules.md— Project rules with dynamic zone (readproject-rules.mdfor template).cursor/rules/memory-bank.mdc— This file's companion: auto-loader wrapper
Step 4 — Fill INDEX.md
The INDEX is the most critical file. It must contain:
- Project overview (name, team, version, status)
- Smart reading strategy decision tree (customized to domains)
- Cache-aware reading order (static before dynamic)
- File map with line counts
- Key technologies summary
Step 5 — Create core and domain files
core/projectbrief.md— Project scope, goals, constraintscore/productContext.md— Product vision, target userscore/active-context.md— Current phase and first taskscore/sprint-tracker.md— Sprint 1 if applicabletech/{domain}-stack.md— One per domain with tech choices and patternsarch/patterns-overview.md— Initial architecture patternsarch/decisions-summary.md— ADR-001: initial architecture
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.
- 9d ago First seen · 82 lines · 36 tokens per session scan A ce8947653d2d
create-memory-bank is a cursor rule published in the GitHub repository mehmetdemirci/comb-ai (13 stars, last pushed 6mo ago), licensed MIT. It adds 36 tokens to every session and 694 once invoked, about $0.0002 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.
Other cursor rules, from other repositories
app-router-patterns
Next.js 14+ App Router patterns — Server Components, Client Components, Route Handlers, Server Actions, and metadata API.
cursor-memory
Shokunin memory system: persistent context across sessions.
aictx
Cursor rule "aictx" from oldskultxo/aictx, covering aictx, priority model, default loop, inspection and advanced tools and what to record.
react-composition-patterns
React composition patterns for scalable component architecture including compound components, context providers, state management, boolean prop alternatives, render props vs children, and React 19 API changes. Use when refactoring components with many boolean props, building component libraries, designing flexible…
react-rerender-optimization
React re-render optimization patterns including memo, useMemo, useCallback, useState, useRef, and derived state. Use when components re-render too often, dealing with stale closures, dependency arrays, performance issues, or optimizing slow components.
160-workflow-orchestration
WHEN executing tasks MANAGE workflow processes for consistent memory operations.