AI & Machine LearningTechnologyPlatinum

Verify LLM output, check if citations are real, or evaluate whether claims are supported by credible evidence.

Source Verification Expert

SIFT Method, CRAAP Test, Citation Verification, Hallucination Detection

intermediate

Best for

▸Evaluating source quality and credibility across domains
▸Detecting LLM hallucinations and fabricated citations
▸SIFT and CRAAP-based source triage
▸Citation existence and accuracy verification

What you'll get

▸Source verification report for 8 cited papers with authenticity status and DOI validation results
▸Citation integrity audit flagging 3 fabricated references and 2 misattributed claims in a draft paper
▸Cross-reference analysis verifying statistical claims against original study data with confidence levels
▸Source quality assessment ranking 15 references by evidence hierarchy (RCT > cohort > case report)

Expects

Content to verify: LLM output, article, research summary, or specific claims with cited sources.

Returns

A structured verification report with claim-level confidence ratings, source quality assessments, red flags, and recommendations.

What's inside

“You are a Source Verification Expert specializing in evaluating source credibility, detecting fabricated citations, and assessing the reliability of claims in any domain. You combine established information literacy frameworks (SIFT, CRAAP) with knowledge of LLM-specific hallucination patterns to pr...”

Covers

The Scale of the ProblemSource ClassificationThe SIFT Method (Caulfield, 2019)The CRAAP Test (Blakeslee, 2004)Lateral vs Vertical Reading (Wineburg & McGrew, 2019)Source Hierarchy by DomainCommon Synthesis ErrorsTriangulation (Denzin, 1978)Conflict of Interest DetectionStep 1: Claim Extraction+5 more

Not designed for ↓

×The research process itself (use deep-research-strategist)
×RAG architecture design
×SEO or keyword analysis
×Academic writing or manuscript preparation
×News media editorial decisions

SupaScore

89.4▼

Research Quality (15%)

9.1

Prompt Engineering (25%)

8.9

Practical Utility (15%)

9.2

Completeness (10%)

8.6

User Satisfaction (20%)

8.8

Decision Usefulness (15%)

Evidence Policy

Standard: no explicit evidence policy.

fact-checkingsource-verificationcitation-checkhallucination-detectionSIFTCRAAPevidence-evaluationtriangulation

Research Foundation: 8 sources (4 paper, 3 industry frameworks, 1 books)

This skill was developed through independent research and synthesis. SupaSkills is not affiliated with or endorsed by any cited author or organisation.

Version History

v1.0.03/14/2026