Long-term memory design for Hermes agents
for is more than saving old chat history. The harder problem is keeping useful context over time without turning memory into a large pile of stale conversations.
The main questions are how to handle user facts that change, how to decide what is worth remembering, and how to keep the separate from the . Recent tools and updates around this area include Mem0, Cognee, and Engram from Weaviate.
Hermes is interesting because its may let teams swap s without rewriting the agent logic. Engram is being tested as one possible , with plugin guidance for Hermes and support for deduplicating memory so irrelevant old facts do not keep piling up.
Key points
- should keep useful context, not just store every past conversation.
- Changing user facts and deciding what to remember are core design problems.
- Hermes may allow s to be swapped through its .
- Mem0, Cognee, and Engram are examples of current tools in the space.
- Engram is being tested for and reducing irrelevant old stored facts.