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/HomenShum/NodeBenchAIWrote 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/homenshum/nodebenchai/nodebench)<a href="https://agentmods.dev/rules/homenshum/nodebenchai/nodebench"><img src="https://agentmods.dev/badge/rules/homenshum/nodebenchai/nodebench/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/homenshum/nodebenchai/nodebench"><img src="https://agentmods.dev/badge/rules/homenshum/nodebenchai/nodebench.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 | $0.00000 | $0.01907 |
| Opus 5.5 | $0.00000 | $0.00763 |
| Sonnet 5.5 | $0.00000 | $0.00381 |
| Haiku 4.5 | $0.00000 | $0.00191 |
Grade A, and why
nodebench 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 1mo 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 — 224 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementation Guidelines
1. Code Reuse & Existing Components
- MANDATORY: Before creating any new component, search the codebase for existing similar components and reuse them
- Leverage existing patterns:
2. Architecture & File Organization
- Follow the existing hierarchical domain-organized structure:
src/features/editor/ # Editor-specific features ├── components/ # React components ├── hooks/ # Custom hooks └── utils/ # Utility functions src/features/agents/ # Agent-specific features └── components/FastAgentPanel/ # Fast Agent UI components convex/domains/ # Backend domain logic ├── documents/ # Document operations └── agents/ # Agent operations - Group related functionality together (hooks with hooks, components with components)
- Keep backend (Convex) and frontend (React) clearly separated
3. Documentation & API Verification
- BEFORE implementing any library/framework feature:
- Use Context7 MCP to fetch the latest official documentation
- Verify API signatures, parameter types, and return values
- Check for version-specific changes or deprecations
- For external integrations (Convex Agent, BlockNote, TipTap, RTVI):
- Use web search to find latest best practices
- Cross-reference with existing usage in the codebase
- Ensure compatibility with current versions in package.json
4. The Standard Development Cycle (Request → Ship)
Follow this exact loop for every task to ensure stability and transparency:
A. Analyze & Plan (Request)
- Search First: Use
grepor file reads to locate relevant files before coding. - Context Check: Identify if the change affects multiple views (e.g., Does this Sidebar change affect both Research and Workspace modes?).
B. Implement
- Write the code following the Architecture and Reuse rules above.
- Apply changes iteratively (e.g., create the component → export it → integrate it).
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.
- 1mo ago First seen · 224 lines · 0 tokens per session scan A a68c7f982589
nodebench is a cursor rule published in the GitHub repository HomenShum/NodeBenchAI (15 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,907 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 2026-09-06.
Other cursor rules, from other repositories
design-patterns
Design patterns (GoF) and file size conventions for sweagent TypeScript codebase.
database
Cursor rule "database" from ItamarZand88/awesome-agent-conventions, covering database best practices, prisma setup, prisma models, prisma queries and supabase setup.
cis-innovation-oracle
CIS Agent — Victor (Disruptive Innovation Oracle).
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.