Agents Lookup

Use cases / Decision models (Jev)

Check submitted copy or web pages for AI-writing tells

Run guest posts, freelancer copy or a site's pages through Jev's checks for common AI-writing tells before you accept or publish them.

DifficultyBeginner
Setup time1 hour
CategoryDecision models (Jev)

Author's figure: Author's figure: 35 tells, one check at 243 ms, $0.00015 (Jon Kraayenbrink). Reported by the builder, not measured by us.

The use case

Editors, agencies and site owners receive copy from freelancers or guest writers and want to know if it reads like unedited AI output before they pay or publish. These tools check text or a whole page against a list of known tells, from filler phrasing to fake testimonials and stock layouts, and give a score with the tells found. The editor uses the list to ask for a rewrite. A high score is not proof that someone used AI, so it should start a conversation with the writer, never decide payment or a ban by itself.

Example workflow

  1. Paste a draft or a URL into the checker
  2. Fetch page text with Firecrawl if checking a site
  3. Ask Jev Noul questions for each tell on the list
  4. Show the score and the tells found to the editor
  5. The editor decides on edits and talks to the writer

Tools

Risks and guardrails

Sources

Written by Agents Lookup. Credits and official pages for the tools this use case names: