What does ruflo-agentdb do?

It lets an agent store memories and find related past information fast using vector search, with HNSW search, RuVector embeddings, and causal graphs.

What kind of thing is it?

It is a plugin providing AgentDB memory controllers for agent memory and retrieval.

ruflo-agentdb

PackSearch

ruflo-agentdb is a plugin that gives an agent a persistent memory backed by AgentDB memory controllers. It stores memories as embeddings and uses HNSW vector search, RuVector embeddings, and causal graphs, so the agent can quickly find related past information when working on a task.

Unavailable. Delivery for this kind is on the roadmap — not serving yet.

Have an agent setup that supports plugins.

What your AI can do with it

Stores agent memories using AgentDB memory controllers
Runs HNSW vector search over stored memories
Generates RuVector embeddings for stored content
Builds causal graphs linking related memories
Retrieves related past information quickly during tasks

Getting started

Have an agent setup that supports plugins.
Add the ruflo-agentdb plugin to it.
Configure the AgentDB memory controllers as the memory backend.
Start a session and let the agent store and search memories.

Signals

GitHub stars: 73k
Forks: 9k
Last commit: Sep 2026

Questions

What does ruflo-agentdb do?: It lets an agent store memories and find related past information fast using vector search, with HNSW search, RuVector embeddings, and causal graphs.
What kind of thing is it?: It is a plugin providing AgentDB memory controllers for agent memory and retrieval.

Advanced

Item type: plugin
Key: ruvnet-ruflo-ruflo-agentdb
Source: github.com/ruvnet/ruflo

github.com/ruvnet/ruflo