FalkorDB

FalkorDB is a Redis-based graph database that supports the property graph model and the OpenCypher query language, includes a built-in graph algorithm module, and stores graph data as sparse matrices.

Cypher is a declarative query language designed by Neo4j for property graph databases, and FalkorDB is compatible with a subset of its syntax.

FalkorDB is an open-source graph database compatible with the Redis protocol and the openCypher query language.

OntoGr runs on top of FalkorDB or Neo4j, and can optionally use pgvector or Postgres; a background consolidation task handles deduplication, summarization, scoring, and archiving; the LLM and embedder are bring-your-own.

Graphiti uses the Apache-2.0 license, supports Neo4j, FalkorDB, or Amazon Neptune, provides a temporal knowledge graph with validity intervals on both entities and relationships, and can query what was true on a given day; it is delivered as an engine rather than a service, requiring you to build identity, authentication, queues, backups, and application APIs yourself.

Graphiti is Zep's open-source temporal graph engine, supporting time-validity intervals for entities and relationships, able to answer "what was true at a given point in time", requires pairing with a graph database (such as Neo4j, FalkorDB, Neptune), and provides a Server and an MCP adapter for self-hosting.

VALOGRAPH's graph model uses the core node types: Player, Agent, Skill, Weapon, Map, Match, and Memory.

VALOGRAPH uses the edge (Agent)-[:HAS_SKILL]->(Skill) to connect agents with their skills, (Player)-[:PLAYS]->(Agent) and (Player)-[:PLAYED]->(Match) to track compositions and match history, and (Player)-[:REMEMBERS]->(Memory) to store persistent AI coach insights across sessions.

VALOGRAPH executes parameterized OpenCypher queries with a response time of about 1.36 milliseconds.

FalkorDB's max flow algorithm rejects parallel edges of the selected relationship type anywhere in the graph, including edges unreachable from the source node, causing maxFlow execution to fail.

In the removal test, all six attack scenarios were blocked or handled when graph context was available, while the control server returned "no opinion" for all six scenarios when the graph was unavailable; the author explicitly states that this is not a general security benchmark and that six scenarios are insufficient to establish a general detection rate.

VIGIL's core graph model is (:Session)-[:MADE]->(:Call)-[:READ]->(:Value) and (:Call)-[:ARG_CONTAINS]->(:Value), and values matching secret key rules are hashed before being stored on Value nodes, with the keys themselves never entering the graph.

The expression RETURN 'ab/c' STARTS WITH 'ab' + '/' returns "true/" on FalkorDB 4.22.1 and returns true on 6.0.2; in a WHERE clause the old version's behavior may cause a type mismatch that makes the query fail, and adding explicit parentheses fixes it.

The VIGIL project contains 68 decision-related queries, all validated on FalkorDB 6.0.2, with a median graph database time per decision of 0.71 milliseconds and a p99 of 2.69 milliseconds.

VIGIL is a runtime firewall backed by the graph database FalkorDB, deployed in front of AI Agent tools, and can serve as a PreToolUse hook for Claude Code or run as an MCP server.

VIGIL uses the max-flow min-cut theorem to turn security decisions into min-cut cost computation, assigning capacity 2 to argument edges and capacity 2n−1 to the call edge (n being the number of arguments), with the max flow return value directly encoding the decision: 0 for allow, 2k for rewrite, 2n−1 for block.

In FalkorDB 6.0.2, query patterns where a procedure call is immediately followed by an aggregation and then a subsequent traversal may silently return empty rows, whereas the same query returns the expected results on 4.22.1.

FalkorDB 6.0.2's algo.maxFlow procedure accepts only a single relationship type, and passing multiple types raises an error; the behavior is the same on 4.22.1.

When a Groq API key is available, Groq can drive the tool loop; when not, deterministic replay executes the same workflow against FalkorDB.

BatchZero is a Graph Hacks 2026 project built for the FalkorDB track, using synthetic data to demonstrate an outbreak investigation workflow; it is not a real public health system.