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.
npx agentmods add rules/forsonny/enhanced-cursor-memory-bank-system/003_mode_definitionsgit clone --depth 1 https://github.com/forsonny/Enhanced-Cursor-Memory-Bank-SystemWhat 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 | $0.00000 | $0.03235 |
| Opus 5 | $0.00000 | $0.01618 |
| Sonnet 5 | $0.00000 | $0.00647 |
| Haiku 4.5 | $0.00000 | $0.00324 |
Grade A, and why
003_mode_definitions 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.
How it starts
The opening of the file, as written. The whole thing — 478 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Operational Mode Definitions
This rule defines distinct operational modes that change how I interact with the codebase and memory system. Each mode has specialized memory behaviors optimized for different phases of development.
Mode Switching Protocol
Mode switching requires two steps:
- UI Selection: First, manually select the mode from Cursor's UI dropdown menu
- Mode Confirmation: Then, signal your intent to the AI:
/mode <mode_name>
Example: Select "THINK" in the UI dropdown, then type /mode think to confirm
I will verify you're in the correct mode before proceeding.
Mode Overview
graph TD
T[THINK] -->|Exploration Complete| P[PLAN]
P -->|Plan Approved| I[IMPLEMENT]
I -->|Implementation Complete| R[REVIEW]
R -->|Issues Found| T
T -->|Small Change| I
R -->|Approved| D[DOCUMENT]
D -->|New Feature| T
style T fill:#f9f,stroke:#333,stroke-width:2px
style P fill:#bbf,stroke:#333,stroke-width:2px
style I fill:#bfb,stroke:#333,stroke-width:2px
style R fill:#fbb,stroke:#333,stroke-width:2px
style D fill:#ffb,stroke:#333,stroke-width:2px
Mode Definitions
THINK Mode
Purpose: Explore problems, research solutions, and understand requirements before implementation.
Memory Behavior:
- Requests access to architecture.md and patterns.md
- Suggests updating working_decisions.md with pros/cons of approaches
- Helps record research findings in session_notes.md
- Does not modify production code
Commands:
/think explore <topic>- Start guided exploration of a topic/think compare <options>- Analyze trade-offs between approaches/think research <query>- Research and document findings on a topic
Memory Files:
- Primary: working_decisions.md, session_notes.md
- Reference: architecture.md, patterns.md, project_brief.md
PLAN Mode
Purpose: Develop detailed implementation plans without making code changes.
Memory Behavior:
- Requests access to current_context.md and patterns.md
- Helps create structured implementation plans
- Suggests updates to current_context.md with task breakdown
- Validates plans against patterns.md and architecture.md
- Helps record planning decisions in decisions.md when finalized
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 · 478 lines · 3,235 tokens per session scan A fc4ea7151dc4
003_mode_definitions is a cursor rule published in the GitHub repository forsonny/Enhanced-Cursor-Memory-Bank-System (35 stars, last pushed 1y ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,235 tokens. 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
agentic-clean-code
Language-agnostic agent-oriented clean code: file size limits, grep-friendly names, provenance comments, testability, and error context. Apply when writing or reviewing code in any stack.
yaml-config
Rules for YAML / config / infra files (Docker, K8s, Terraform, GitHub Actions, etc.).
agent-efficiency
Agent-mode tool-call efficiency. Cuts the largest hidden cost in modern AI IDEs — wasted tool calls and oversized context windows.
data-science
Rules for data science, ML, and Jupyter notebook work.
mobile
Rules for mobile development (Swift/iOS, Kotlin Android, React Native, Flutter).
nextjs
Rules for Next.js projects (App Router and Pages Router).