ClaimTrace – open-source engine that traces public claims to preserved evidence
ClaimTrace is an open-source engine designed to trace public claims to preserved evidence without labeling individuals as frauds. It focuses on assessing the support of public claims through preserved evidence while maintaining a neutral stance. The engine separates claims, evidence, and conclusions to ensure clarity and credibility in its reporting.
- ▪ClaimTrace captures public sources and preserves evidence with hashes and metadata.
- ▪The engine is built to support claim-review workflows without making consumer verdicts.
- ▪It does not classify individuals as frauds or scammers, focusing instead on evidence support.
Opening excerpt (first ~120 words) tap to expand
ClaimTrace ClaimTrace is an open-source engine that traces public claims to preserved evidence and refuses to call anyone a fraud. It asks one question: How well is a public claim supported by preserved public evidence? ClaimTrace captures public sources, preserves readable evidence with hashes and metadata, links claim candidates back to source excerpts, and produces decomposed reports where credibility strength, credibility risk, and evidence coverage stay separate. It is built to support claim-review workflows without turning the engine into a consumer verdict machine. Quickstart cp .env.example .env docker compose up --build The API will be available at http://localhost:8000.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.