Get clear.
Then get paid.
A rushed client brief hides expensive unknowns. Check what's missing in seconds, then ask smarter questions with a tiny open-weight model running on your device.
Paste a brief. See the gaps.
Sample details are fictional. Enter a real brief only if you're comfortable processing it in your own browser.
⌁ Heuristic checks run offline. AI weights download once only when you request it; your brief is not sent to a hosted AI API.
READY FOR INSPECTION
Answers before assumptions.
Your scope report will appear here after you run a check.
Method note: the first scan checks for keyword signals, not actual contractual sufficiency. The AI model generates text and may be inaccurate. Always verify scope and commercial terms with the client.
No cloud.
No guessing.
The boring but costly questions often fall through the cracks. ScopeLens combines transparent checks with optional private model inference.
Transparent first pass
Eight visible topics. Each gap has a fixed, editable client question. The heuristic scan requires no downloads or API keys.
Open-weight second opinion
SmolLM2-135M-Instruct runs with Transformers.js on your device after the model is downloaded. The generated questions are suggestions, not fact checks.
Take it to your workflow
Export Markdown notes, revise the scope with your customer, then use your own quoting or invoicing process. We store no brief on our servers.
INDEPENDENT. OPEN-SOURCE. ADAPTABLE.
The most expensive word
in a brief is “assume.”
This prototype is free and open source. If you want a tailored checklist for your agency's workflow, get in touch for a scoped discussion—no payment flow here.
The linked form is handled by third-party FormSubmit with explicit consent; unlike ScopeLens itself, contact form details leave your device. / 문의 양식은 동의 후 제3자 FormSubmit으로 전송됩니다.
Public GitHub inquiry ↗Public GitHub issue; account required. No confidential details, personal information, passwords, or payments.