Product docs
RugTrace Brain docs
RugTrace Brain is positioned as post-rug investigation plus persistent scam memory, not a pre-buy token scanner.
RugTrace Brain does not recover funds or make legal accusations. It reconstructs public on-chain fund flow and labels suspicious graph patterns.
What RugTrace Brain does
- - Post-rug investigation, not a pre-buy token scanner.
- - Reconstructs victim-side and suspect-side public on-chain fund flow.
- - Matches historical cases, wallet behavior, and suspicious graph patterns.
- - Generates AI forensic reports grounded only in graph facts.
How it works
- - Parse a Solana transaction signature.
- - Build victim-side and suspect-side fund-flow graphs.
- - Match against Brain Memory patterns and case memories.
- - Generate an Evidence Table and grounded report.
- - Save new case memory for future investigations.
What Brain Memory means
- - Built-in patterns: transparent rules and feature vectors.
- - Case-library memory: investigative signals generated from curated OSINT cases.
- - DB memory: Memory records saved after app investigations.
- - Wallet profiles: case-related wallet tags and risk features.
Data source badges
- - LIVE HELIUS: data came from live Helius / Solana RPC.
- - FALLBACK FIXTURE: local demo data was used when external APIs were unavailable.
- - PARTIAL LIVE + FALLBACK: live data and fallback data were combined.
Safety and limitations
- - No fund recovery and no recoverability claims.
- - No legal accusations and no real-world identity attribution.
- - Candidate transactions need review.
- - Reports explain existing graph facts and should not be treated as legal conclusions.