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/homenshum/nodebenchai/dogfood_verificationgit clone --depth 1 https://github.com/HomenShum/NodeBenchAIWhat 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 | $0.01407 |
| Opus 5 | $0.00000 | $0.00704 |
| Sonnet 5 | $0.00000 | $0.00281 |
| Haiku 4.5 | $0.00000 | $0.00141 |
Grade A, and why
dogfood_verification 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 2d 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 — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dogfood Verification Protocol — Product Design Review
Every UI change MUST be visually verified in the running app before declaring done. Code-level changes without visual confirmation are unverified assumptions. But verification is not just "does it render?" — it's "does it belong?"
Mindset: Analyst, Not Junior Dev
Guide yourself like an analyst diagnosing root cause, not a junior dev slapping on a bandaid. Before touching code:
- Form a hypothesis — What is actually wrong? "It looks broken" is not a hypothesis. "The date picker generates UTC dates but the data is keyed by local dates" is.
- Trace the render path — Which component ACTUALLY renders on screen? Follow: sidebar item → view key → MainLayout switch → lazy import → component file. Don't assume — the component you grep for may not be the one the user sees.
- Identify the data layer — Is this Convex backend (needs
convex deploy), React frontend (needsvite build), stored DB records (needs migration), or MCP tools (needstsc)? Changing the wrong layer wastes a cycle. - Ask "why" before "what" — If a gauge shows 0%, don't change the label. Ask why it's 0%. Missing data? Wrong query? Unconfigured system? Each has a different fix.
Jony Ive Product Design Critique
After verifying it renders, apply the design critique. Every element must earn its place:
Does it communicate or decorate?
- Every stat must drive a decision. If a number doesn't change behavior, it's decoration — remove it.
- "14 threads" means nothing without context. "14 threads (3 escalating)" drives action.
- Streak counters, "free runs today", total counts without trends — vanity. Remove.
Does it respect the user's time?
- Empty states: "No data yet. Run:
npm run local:sync" gives agency. "Nothing here" wastes time. - Loading states: skeleton shimmer, not blank void. User should never stare at white.
- Error states: what went wrong + what to do about it. Never just "Error".
Does the language earn trust?
- Replace jargon: "Act I/II/III" → "Quick Pulse / Analysis / Deep Dive"
- Replace internal labels: "Orchestrator" → "Agent", "Trajectory" → "History"
- Test: would a non-technical person understand this label in 2 seconds?
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.
- 2d ago First seen · 101 lines · 0 tokens per session scan A 0ec7965b7a26
dogfood_verification is a cursor rule published in the GitHub repository HomenShum/NodeBenchAI (14 stars, last pushed 19d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,407 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
api-property-optionality-hygiene
Fix ApiProperty/ApiPropertyOptional optionality mismatches in DTO files; use for scheduled batch fixes or DTO edits.
java-springboot-jpa-cursorrules-prompt-file
description: "Cursor rules for Java development with Springboot and JPA integration." globs: / alwaysApply: false.
conservative-file-creation
Check for existing files before creating. Be conservative about new files and code.
secure-dev-rust
These rules apply to all Rust code in the repository and aim to prevent common security risks through disciplined use of memory safety, input validation, error handling, and safe APIs.
design-patterns
Design patterns (GoF) and file size conventions for sweagent TypeScript codebase.
cursorrules
ALWAYS start your session by reading AGENTS.md and .memory/wiki/hot.md to get project context before suggesting code or answering questions.