URL & Source Credibility Assessment

Fast AI analysis of domain reputation, publication history, and content credibility — no file upload required.

The URL & Source Credibility Assessment pipeline is Crowdee's fastest and most lightweight verification option. It assesses the credibility of a URL and its source without requiring any file upload. Ideal for real-time editorial workflows, bulk URL screening, and rapid first-pass triage.

Unlike pipelines that operate on uploaded media, this pipeline works entirely from the URL you supply in the context object. The AI fetches and analyses the target page alongside signals about the domain's history, reputation, and known classifications — delivering a credibility verdict in approximately 2 minutes.

Pipeline Details

Property	Value
Pipeline ID	`verify-url-credibility`
Tier	3
Estimated Duration	~2 min
Credit Cost	500 credits per run
Crowd Stages	No
Files Required	No files required

Required Inputs

Files

No files required.

Context Keys

Key	Required	Description
`url`	Required	The URL to assess for credibility.
`claimed_topic`	Optional	Topic or subject matter of the content at the URL.
`claimed_date`	Optional	Claimed publication date of the URL's content.

Stages

Domain Analysis (AI)

Evaluates the domain hosting the URL. The AI examines domain age and registration details, publication history, known bias classifications from third-party indices, and presence in disinformation or state-affiliated media databases. A recently registered domain with no publication history publishing high-stakes claims is a meaningful credibility signal, even before the content itself is read.

Content Context (AI)

Analyses the content at the URL (if accessible): headline, author, byline, cited sources, and consistency with established facts. The AI checks whether the claimed topic and publication date match the page content, and whether the writing style, sourcing practices, and factual claims are consistent with credible journalism.

Verdict (AI Synthesis)

Aggregates domain and content signals into a single credibility verdict with a confidence score and a plain-language summary explaining which factors most influenced the outcome.

Starting a Run

curl -X POST https://api.crowdee.ai/v2/projects/{projectId}/verification-runs \
  -H "X-API-Key: crw_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "pipeline_id": "verify-url-credibility",
    "context": {
      "url": "https://example-news.com/article/2024-11-15",
      "claimed_topic": "election results",
      "claimed_date": "2024-11-15"
    }
  }'

Example Response

{
  "id": "run_01j9x...",
  "status": "completed",
  "verdict": "inconclusive",
  "confidence": 55,
  "pipeline_id": "verify-url-credibility",
  "scorecard": {
    "domain_age": "warn",
    "bias_index_match": "unknown",
    "content_consistency": "pass",
    "claimed_date_match": "pass"
  },
  "explanation": "Domain was registered less than 6 months ago with limited publication history. Content is internally consistent and the claimed date matches the page, but the domain's youth and absence from established indices lower overall confidence."
}

This is Crowdee's fastest pipeline — typically completes in under 2 minutes. No file upload is needed; pass the URL in the context object. Omit file_ids entirely.

The verdict reflects source credibility, not factual accuracy. For claim-level fact-checking, use verify-news-article.