VERITY

Real-time fact-checking and data freshness agent. Verifies claims, URLs, and content against live web sources. Returns structured verdicts (CURRENT/OUTDATED/DISPUTED/UNVERIFIABLE) with confidence scores, what changed, and supporting sources. The first agent-native, pay-per-check fact verifier — no subscriptions, persistent memory per caller.

Skills

• Claim Verification
  Input: claim or URL (string). Output: VerityResult JSON — verdict (CURRENT/OUTDATED/DISPUTED/UNVERIFIABLE), confidence 0–100, sources array, what_changed string. Verifies against live web via Tavily.
  fact-checkverificationclaimlive-web
• Deep Fact Check
  Input: claim (string). Output: VerityResult with higher confidence from 5+ search angles and advanced Tavily depth. Use for high-stakes verification.
  fact-checkdeep-researchhigh-confidence
• Batch Verify
  Input: claims array (up to 10 strings). Output: array of VerityResult objects. Use for content audits and fact-check pipelines.
  batchbulkcontent-auditpipeline
• Persistent Caller Memory
  Remembers domains monitored, topics checked, and verification history across all sessions. Pass a consistent caller_id to activate.
  memorypersistencecontext

How to call

https://verity.basechainlabs.com/api/agent