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/vibestackdev/vibe-stack/verify-before-usegit clone --depth 1 https://github.com/vibestackdev/vibe-stackWhat 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 | $0.00541 | $0.00541 |
| Opus 5 | $0.00270 | $0.00270 |
| Sonnet 5 | $0.00108 | $0.00108 |
| Haiku 4.5 | $0.00054 | $0.00054 |
Grade A, and why
verify-before-use 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify Before Use
RULE 1: Run Type Checker After Every Code Generation
After generating or modifying TypeScript files, ALWAYS run the type checker before declaring the task complete. Do not assume code is correct — verify it.
npx tsc --noEmit
If types fail, fix ALL errors before moving on. Do NOT leave type errors for the developer to fix manually.
RULE 2: Re-Read Before Modifying
Before modifying ANY existing file, re-read it first. Do NOT generate code based on assumptions about what the file contains.
Common failures from skipping this step:
- Importing functions that were renamed or deleted
- Overwriting changes made in a previous turn
- Adding duplicate imports or declarations
- Breaking existing patterns by not matching the file's style
RULE 3: Verify Database Schema Before Queries
Before writing Supabase queries, verify the table structure by checking
the migration files in supabase/migrations/.
Do NOT assume column names, types, or relationships. Common hallucinations:
- Using
user_idwhen the column isauthor_id - Assuming a
statuscolumn exists when it doesn't - Using wrong relationship names in
.select('*, profile(*)')
RULE 4: Test Route Patterns Before Declaring Done
After creating new API routes or pages, verify:
- The file path matches the expected URL pattern
- The HTTP method matches the use case (GET for reads, POST for mutations)
- The response shape matches what the client expects
app/api/posts/route.ts → GET /api/posts, POST /api/posts
app/api/posts/[id]/route.ts → GET /api/posts/123, DELETE /api/posts/123
app/dashboard/page.tsx → /dashboard
app/posts/[slug]/page.tsx → /posts/my-first-post
RULE 5: Verify Environment Variables Exist
Before using process.env.SOME_KEY, verify it exists in .env.example.
If the variable is new, ADD it to .env.example with a placeholder value
and document it in the setup instructions.
NEVER put secrets in NEXT_PUBLIC_ variables — they are exposed to the browser.
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 · 71 lines · 541 tokens per session scan A aca76e443f43
verify-before-use is a cursor rule published in the GitHub repository vibestackdev/vibe-stack (7 stars, last pushed 2d ago), licensed MIT. It adds 541 tokens to every session, about $0.0027 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 2026-08-31.
Other cursor rules, from other repositories
cursor
SEO audit and implementation for Next.js, Remix, and React SPAs. Auto-attached to page, layout, and route files. Covers meta tags, Open Graph, Twitter cards, JSON-LD, sitemaps, robots.txt, i18n/hreflang, and image SEO.
diagnose
name: Issue Re-Diagnosis and Fix Template description: Template for AI to re-diagnose and fix issues that weren't resolved previously. globs.
always-on
Cursor rule "always-on" from jjcall/vibeship-boilerplate, covering core persona, how you work in this project, how you verify and validate, when you should ask for feedback or approval and how you communicate.
request
name: Issue Re-Diagnosis and Fix Template description: Template for AI to re-diagnose and fix issues that weren't resolved previously. globs.
upgrade-tests
// ✅ CORRECT: Upgrade package pattern await using ctx = testReactResource(appRouter, { server: { // server configuration }, client(opts) { return { links: [ httpLink({ url: opts.httpUrl, // client configuration }), ], }; }, }).
react-query-tests
// ✅ CORRECT: Legacy React Query pattern const ctx = konn() .beforeEach(() => createAppRouter()) .afterEach((ctx) => ctx?.close?.()) .done().