TASKSCAN · rubric v1
Which of your tasks can an agent already do?
Paste your tasks, one per line. Move four sliders for each one and see the keep / delegate / automate split, computed in your browser with the reasoning shown. No account, no payment, no network call.
1. Paste your tasks
Plain words are enough. Do not paste confidential, customer or personal data: this page runs locally, but anything you later send for an AI report leaves your browser.
2. Your keep / delegate / automate split
Every task starts at 3 on all four sliders — a starting guess, not a score. Adjust them to match your own work and the split updates instantly.
0 / 100 overall exposure
Overall exposure bar
Keep 0
34 or below: the value is in the judgement, the presence or the relationship.
Delegate 0
35 to 64: an agent can draft or prepare it; a person still decides.
Automate 0
65 and above: an agent could do the task with a person checking the result.
Drag a card to another lane, or focus a card and press Enter to pick it up, left or right arrow to choose a lane, Enter to drop, Escape to cancel.
The same data as a table
| Task | D | C | E | L | Physical | Exposure | Lane |
|---|
AI report coming soon, the free sorter works now
3. Want the model to score your real list?
Your split above is your own arithmetic. An AI model scores the same rubric on your real task list, writes a one-line reason for each task, and produces a 90-day plan. One payment, no account, no subscription.
Personal report: $9 · the price is shown before you click, and nothing is charged if the report fails to generate.
Refund line: if your result fails to generate or is unusable, contact us within 7 days for a refund.
Your AI report preview
Your full report
Share card
Right-click or long-press to save the image. It shows the overall score only, never your task text.
Anyone holding your result link can read, export or delete it. Your task text is deleted from our server after 30 days; the report stays at your link.
Something went wrong
What we send and to whom
The split above is computed in your browser and nothing is sent anywhere. If you ask for the AI report, we send your task list (at most 40 tasks) to the AI providers named in the product configuration — Google Gemini, then Groq, then Cloudflare Workers AI — and store the result at your private link.
Free AI tiers may allow the provider to use submitted content to improve its services, so do not paste confidential or customer data. We keep the result at your private link, delete the text you entered after 30 days, and you can delete your result at any time. Anyone who has your link can read your result.
How the score is worked out
Four numbers from 0 to 5: D digital end-to-end (0 = needs hands or a place, 5 = fully on a computer), C output checkable (0 = pure judgement, 5 = a test or reconciliation exists), E error cost (0 = trivial, 5 = severe), L liability or licensure burden (0 = none, 5 = personal legal liability).
exposure = 100 * (0.35*D + 0.30*C + 0.20*(5-E) + 0.15*(5-L)) / 5 if the task needs physical presence: exposure = min(exposure, 15)
Lanes: 65 and above Automate, 35 to 64 Delegate, 34 and below Keep. The weights are a v1 opinion, published on the methodology page; the golden set is proposed, pending owner hand-scoring.
This tool gives informational estimates, not professional, legal, financial or career advice. Results depend on what you enter and on data and model limits. We do not guarantee outcomes. Rubric-based estimate, not a prediction. TASKSCAN never says a job will disappear.
Free layer: methodology · most exposed occupations · least exposed occupations · department audit, $149
Privacy · Terms · Refund policy · Found an error? Tell us