citation-verification

Verify academic citations by running multi-source lookups across OpenAlex, Crossref, and Semantic Scholar to confirm existence and metadata accuracy. The Skill extracts references from uploaded documents or pasted lists and returns structured match quality with evidence links.

Check AI-generated references by extracting citations from your manuscript and querying authoritative databases like OpenAlex and Crossref. The verification process assigns match quality scores—EXACT, HIGH, MEDIUM, LOW, or NONE—to flag fabricated or incorrect entries.

Perform batch citation verification by submitting your literature review to cross-database lookups against Semantic Scholar and Crossref. Instead of binary results, the system provides confidence levels and evidence trails, resisting batch pressure and ensuring reproducible match quality scoring.

Yes, you can verify references directly from an uploaded grant proposal or manuscript. The Skill automatically extracts pasted references or processes uploaded files, then checks them against multiple databases to report metadata accuracy and existence with structured confidence scores.

Yes, paper-ladder is a required dependency to run Crossref and OpenAlex citation checks. The Skill relies on this prerequisite component to execute its internal scripts for multi-source verification and structured evidence reporting.

Citation verification returns NEEDS_REVIEW because the core principle is to never give binary yes/no results. Match quality scores—including NEEDS_REVIEW, EXACT, and NONE—report confidence levels and evidence trails, ensuring transparency when database lookups yield partial or ambiguous metadata matches.