INDEPENDENT GUIDES / EVIDENCE-FIRST THINKING

Less guesswork.
More
evidence.

Make every claim easier to check. Explore fact APIs, LLM verification, and citation workflows built around sources—not just confidence.

Practical guides. Inspectable examples. No black-box promises.

evidence-record.jsonEXAMPLE

JSON claim → source → assessment

01 / THE CLAIM

“The Atlas demonstration archive opened in 2021.”

{
  "claim_id": "demo-001",
  "status": "supported",
  "evidence": {
    "source": "example-release-note",
    "passage": "Opened in 2021"
  },
  "scope": "fictional evidence pack"
}
Traceable support. Clearly defined limits.

THE METHOD / 4 CONNECTED STEPS

Don’t stop at
“sounds right.”

A useful fact-checking workflow leaves a trail you can follow. Our proposed method starts with a precise statement and ends with an assessment you can inspect—not just a badge.

Read our evidence principles
01

Extract the exact claim

Preserve who, what, when, and every qualifier that changes the question.

02

Find inspectable evidence

Locate a relevant passage. A familiar source name is not enough.

03

Assess the relationship

Distinguish support, contradiction, and evidence that cannot resolve the claim.

04

Keep the trail connected

Record the scope and source version, then review meaningful changes.

THE QUESTIONABLE SHORTCUT

“The model is confident, so it must be correct.”

THE BETTER QUESTION

“Which source supports this exact claim—and what does it leave unresolved?”

A CLOSER LOOK / LOCAL EXAMPLES

One source.
Three different outcomes.

Change the claim, not the evidence. These fixed examples show why support, contradiction, and missing information belong in different categories.

Inspect the JSON examples

ILLUSTRATIVE DATA ONLY. Atlas is fictional.
No statement is submitted for verification.

The claim

The Atlas demonstration archive opened in 2021.

“The Atlas demonstration archive opened in 2021. It accepts JSON exports.”
Supported

The stated opening year matches the supplied example passage. This assessment applies only to this fictional evidence pack.

LOCAL, FICTIONAL EXAMPLE · No live checks or external requests.

THE EVIDENCE JOURNAL

Better questions.
Stronger source habits.

All 10 fieldnotes

USEFUL DISTINCTIONS

Start with
the right question.

Clear boundaries make better tools.
Here is what the site does—and does not do.

What is a fact API?
The term can refer to structured data, published fact-check retrieval, or claim assessment. Our Fact API guide separates these jobs so a returned record is not mistaken for a completed verification.
Does FactAPI.com provide a live fact-checking service?
No. FactAPI.com is an independent resource library with practical guides and downloadable local examples. There are no accounts, API keys, paid plans, or live verification endpoints on this site.
Can an LLM check its own answers?
A model can propose an assessment, but its confidence is not independent evidence. Our LLM workflow connects claims to passages and keeps unresolved cases visible for review.
Is a real citation enough to support a claim?
No. Citation identity and claim support are different checks. The citation-checker guide explains how to inspect the passage, context, and scope of a reference.
Is FactAPI.com affiliated with Snopes?
No. References to Snopes are independent editorial discussion. We do not claim an official Snopes API or a data partnership. Read the online fact-checking guide for careful use of published reviews.
What does a source watch monitor?
Our proposed workflow connects source changes, correction notices, and retractions with the claims that depend on them. The source-watch guide describes the design; this site does not run live monitoring or send alerts.

Follow the claim. Find the evidence.

Go from a better question to a clearer evidence trail—one fieldnote at a time.

Explore the journal