Resume Bullets Need Editing, Not Invented Metrics

A full resume generator can make weak evidence sound impressive by quietly adding scale, ownership, tools, or percentages. That is polished fiction, not career help.

That distinction is why Resume Bullet Optimizer exists as a focused tool rather than another broad prompt box.

What the app actually does

Resume Bullet Optimizer works one level lower. It rewrites supplied bullets into concise, impact-led, and role-aware variants while preserving the underlying evidence. Missing numbers become explicit verification prompts instead of fabricated claims.

The interface loads the live public AI Pass model catalog and uses only exact publicly available model IDs. Authentication stays lazy: anyone can inspect the app first, and AI Pass asks for sign-in only when generation begins.

Why deterministic rules matter

The useful part of these products is not a longer prompt. It is deciding which tasks belong to AI and which must remain deterministic. AI can interpret messy language and draft a useful structure. Browser code can enforce totals, allowed strategy codes, citation matching, output sections, history limits, and safe rendering.

That split makes the app easier to inspect. It also keeps a plausible model answer from quietly becoming an accepted fact.

Private local history

Generated outputs can be copied, downloaded, or shared. Up to twenty results are stored in IndexedDB on the current device. Raw sensitive inputs are not added to history. Each saved output can be opened or deleted, and the full history can be cleared. Raw target titles and industries are not persisted with those saved outputs.

Before using the result

• Compare every variant with what actually happened.
• Add a number only when you can verify it.
• Treat job-description language as a targeting signal, never as proof of experience.

From prompt carousel to product

The starting point was a ten-prompt carousel from @chatgptricks. Instead of reproducing ten generic ChatGPT wrappers, we checked the existing AI Pass catalog, merged overlapping ideas, and narrowed each tool around one job with an explicit evidence boundary.

Open Resume Bullet Optimizer on AI Pass. AI Pass includes $1 in credit on signup, then usage is pay as you go.

---
name: resume-bullet-optimizer
description: Deploy or maintain the self-contained AI Pass Resume Bullet Optimizer Space, including its truth safeguards, bullet-only scope, private output history, tagged streaming protocol, embedded helpers, and deterministic checks.
---

# Resume Bullet Optimizer

Use this package when publishing or updating the Resume Bullet Optimizer Space.

## Operating rules

1. Keep `index.html` as the complete deployable artifact. Do not introduce a build dependency unless the product genuinely requires one.
2. Preserve the literal `PLACEHOLDER_CLIENT_ID` in source templates; replace it only in the deployment copy with the registered AI Pass OAuth client ID.
3. Keep `requireLogin: false`, the visible `data-aipass-button`, eager live model-catalog normalization, and `AiPass.streamText`.
4. Never add `maxTokens` to the streaming request.
5. Do not auth-gate the form or results UI.
6. Never weaken the system prompt’s ban on invented metrics, facts, numbers, tools, responsibilities, ownership, or outcomes.
7. Render missing quantification as a clearly marked verification placeholder, not a claim.
8. Keep the product narrowly focused on evidence-preserving bullet rewrites. Do not expand it into a full resume writer, summary generator, or experience inventor.
9. Treat streamed model text as untrusted. Use `textContent`, text nodes, or equivalent safe rendering; never assign it to `innerHTML`.
10. Populate the model selector only from the intersection between the live catalog and the verified ID allowlist.
11. Keep the private IndexedDB output history with open, delete, and clear-all controls, capped at 20 entries.
12. Keep copy, export, and share actions for generated outputs.
13. Never store raw bullets, job descriptions, resume source material, raw target titles, or raw industries in local history.

## Validation

Run all checks from this directory:

```sh
npm test
npm run build
npm run check
```

All three commands must pass before handoff. The tests exercise the exact helper implementation embedded in the production HTML.