Evally
Start free
AI output quality

Citation checker

Does the cited passage actually support the claim it's attached to?

For research assistants, legal and medical summaries, and any answer with footnotes: checks each claim against the passage it cites and catches the quiet overstatements ('always' where the source says 'often').

Use this judgeCustomize it

Try it live

Real calls to Jev. Pick an example or paste your own.

Examples
42
133
No verdict yet

Pick an example or paste your own input, then run it.

Under the hood

Every question and rule, in the open. Fork it to change any of them.

What it asks

Reads claim and passage, then asks 2 questions in one call.

  1. Q1SupportPick one

    Does the passage support the claim?

    SupportedPartly supportedUnrelatedContradicted
  2. Q2Overstates the sourceYes / no

    Does the claim say something more strongly or more specifically than the passage does? (For example 'always' where the passage says 'often', or an exact figure the passage doesn't give.)

How it decides

Rules run in code, top to bottom. The first one that matches gives the verdict.

  1. 1
    If Support is Supported and Overstates the source ≥ 0.6Partly supportedreview
  2. 2
    Otherwisethe answer to “Support”
  3. ·
    DefaultPartly supported

Call it from your code

Every template runs as-is with an API key, no setup: POST /api/v1/judges/citation-checker/run.

curl -X POST https://tryevally.com/api/v1/judges/citation-checker/run \  -H "Authorization: Bearer $EVALLY_API_KEY" \  -H "Content-Type: application/json" \  -d '{  "input": {    "claim": "Remote work always increases productivity.",    "passage": "In our sample of 1,200 employees, 58% reported being more productive when working remotely, while 22% reported being less productive."  }}'

Seen in the wild

Community builds on Jev that use this pattern. Not affiliated with Evally.

More in AI output quality