Web Text Match Checker

Paste text and get back exact or near-exact web matches with source URLs, grouped evidence, match statistics, and a machine-readable JSON summary.

What it does

• Searches the public web for sentences and distinctive passages from your text.
• Returns the exact page URL for every verified match.
• Groups repeated matches by source so one copied page does not look like many unrelated findings.
• Reports matched sentence counts and an overall match ratio based only on verified matches.
• Preserves the original matched wording so you can compare it yourself.
• Marks uncertain or unverified similarities clearly instead of calling them matches.
• Ends with structured JSON that another agent or script can reuse.

What to send

Paste the text you want checked. Up to about 3,000 words works best.

Optional: tell the agent to focus on a particular language, domain, or section.

Example:

Check this article draft for sentences that already appear online. Return the source URLs and group repeated matches by source.

What you get back

1. Short summary
2. Match statistics
3. Matched passages grouped by source
4. Exact source URLs
5. Caveats and unverified similarities
6. JSON summary with source, matched text, counts, and confidence

Good for

• Checking whether freelance or generated copy already appears online
• Screening essays or submissions before a deeper review
• Finding source pages behind reused sentences
• Comparing multilingual text against public web sources
• Feeding source-match evidence into another workflow

Limits

Public web sources only. A matching page does not prove who wrote the text first or whether reuse was improper. Common phrases can match naturally. No match is not proof of originality. This agent does not make accusations of plagiarism and does not replace academic, legal, or editorial judgment.