1. Prepare the source
Collect the raw text you already have — receipts, notes, messages — with dates and amounts. Remove private details you do not need to share.
proof sheet · free workflow
Turn Headline Options Into a Tested Board With Audience Fit — Free
List 6 headline options and the board scores each on clarity, curiosity, and audience fit using your own criteria.
free path works offline firststarter loads with no network call
12 rows x 90 sec = ~18 min savedmeasured against your own input, not a vague claim
designed result, not prose dumpreceipt / board / packet with hierarchy
What this replaces
scores headlines against the reader's own audience description, not generic clickbait rules The result is a small object you can inspect: it carries a count, a context row, and a next action instead of asking you to trust a paragraph.
Tool note. This workflow uses GratisAI local-first headline board (no key, no account). GratisAI local-first headline board (no key, no account) documentation was checked on 2026-09-13; limits can vary by account. This board beat a generic summary because it keeps source context visible. Runners-up were a generic list generator (too narrow) and a plain summarizer (no checkable output). Use sample data first, then decide what information you are comfortable keeping on this device. Concrete example: CHEM 101 exam Sep 18 2026, 4 units, $412,000 at 6.8% over 30 years, closing June 2027 — four points across three types (proper noun, dollar figure, date, quantity) labeled as sample.
The handoff
2. Ask for strict shape
Prompt the free tool to return only JSON with this shape: {"heading":"...","metric":"...","rows":[{"label":"...","detail":"...","value":"..."}],"next":"..."}. Require it to say "not found" when data is missing.
3. Run the misses here
Paste the same lines into the board below. The companion runs locally and shows a designed result without calling an API.
Prompt shape for external tool: return only JSON in this form — {"heading":"...","metric":"...","rows":[{"label":"...","detail":"...","value":"..."}],"next":"...","quote":"..."}. The board below uses the same visible result shape locally and makes no AI request.
Try the proof sheet board
Load a sample, replace it with your own notes, and inspect the generated object. Sample values are illustrative, not verified facts.
Loading the sample object…
share line
What breaks first
A board that covers only the loudest item
Require one row per source line and check the count before trusting the total. If the count is short, add the missing line.
An answer that sounds plausible but is not in your paste
Keep the evidence phrase and source-grounding request. "Not found" is a useful result that prevents invention.
A low score that feels like a verdict
Treat it as a routing signal for the next seven days, not a prediction. Mark two rows you could not answer and review those sections.
Go deeper without losing the free path
The free board is complete on purpose. The companion app adds a tangible file, history, or handoff only after the result is useful: Printable headline board with 6 options; estimate: 6 headlines x ~2 minutes of manual scoring = ~12 minutes avoided
open the proof sheet companion