AI agents that find money-laundering rings no single bank can see, on FalkorDB: one graph per bank, a consortium graph, echo-proof evidence, signed reports and a hosted MCP server.
-
Updated
Oct 18, 2026 - TypeScript
AI agents that find money-laundering rings no single bank can see, on FalkorDB: one graph per bank, a consortium graph, echo-proof evidence, signed reports and a hosted MCP server.
Graph-native AML (Anti-money Laundering) investigation agents: an Investigator and a Reporter that never talk to each other, only through a FalkorDB graph. Every claim is cited back to exact transaction, account and device nodes.
Theory-of-mind memory for AI agents on FalkorDB: what each agent believes, and what it thinks others believe. Graph Hacks 2026.
A FalkorDB repair lab and MCP tool server for brick models, with counterfactual queries and independently replayable repair proofs.
Trace how a headline moves through the economy on a FalkorDB causal graph
A company brain for an open-source repo on one temporal FalkorDB graph: who knows what, why code is the way it is, what leaves with a departing maintainer, and what-ifs on GRAPH.COPY forks.
Relay: shared, time-aware graph memory for teams of AI agents, built on FalkorDB. The agents never message each other, yet never contradict each other. Live demo: WiFight, an AI ISP support desk. Graph Hacks 2026 · Track 02: Agent Memory & Coordination
Pick the host city your community can actually reach: visa refusal rates, entry rules and routes in a FalkorDB graph
On-call agent that won't blame your deploy without proof. A FalkorDB graph, 4 agents and 8 evidence checks find what really broke production. 81% top-1 on all 360 RCAEval failures. Demo video: https://youtu.be/arIp5REAKmw · Graph Hacks 2026
Graph-native fraud-ring investigator with persistent case memory — FalkorDB as the agent's context layer. WeMakeDevs Graph Hacks 2026 (Tracks 1+2).
A company brain on FalkorDB: one graph per product, nine read-only Cypher tools, tested on 1,203 held-out HERB questions (Graph Hacks 2026, Track 3)
A Company Brain on FalkorDB: reads docs, tickets and chat into a provenance graph and answers who owns, who knows and why, with quoted sources, or refuses to guess. Graph Hacks 2026, Company Brain track.
BlastRadius: an AI agent that finds the fixes that cut real attack paths in a FalkorDB graph, and acts on them within graph-decided limits. Graph Hacks (WeMakeDevs x FalkorDB) Track 1.
Ripple: a FalkorDB company brain for policy change impact, evidence-backed repairs, controlled publishing and persistent agent context. Graph Hacks Track 03.
Ask why a decision was made: a FalkorDB company brain over the IETF TLS working group's public record. Every answer is a graph path to a public link. Graph Hacks 2026, Track 03.
FalkorDB-backed release agent that remembers its failures. Graph Hacks submission — Track 2 (agent memory & coordination).
OwnerGraph: who may approve, who actually decides, and why - Kubernetes governance as one FalkorDB graph (Graph Hacks 2026)
Find the last lift before it fails. Every MBTA station's step-free routes as a flow network in FalkorDB: certified max-flow redundancy, live lift closures, what-ifs and step-free reroutes, served to AI agents over MCP. Unofficial; open data.
Trace a drug shortage to the plant that makes it, find other formulary drugs made there, and pick substitutes made somewhere else. Built on a FalkorDB graph of openFDA, DailyMed and FDA DMF data.
Incident root-cause agents on a FalkorDB temporal graph: MCP graph tools, edge-cited evidence, time-aware queries, and human-confirmed memory. Graph Hacks 2026, Tracks 01 + 02.
To associate your repository with the graph-hacks topic, visit your repo's landing page and select "manage topics."