Persistent AI Memory
A session ends. The project keeps moving. Persistent memory gives connected agents a place to save decisions, constraints, and lessons for later retrieval.
Read overviewResourcesTechnical
Technical
resources
Memory concepts, governance workflows, and deployment considerations for agents that need useful context to persist.
📚Core Concepts
AI Agent Memory Patterns
Agents need different kinds of context at different points in a workflow. A project decision may need to persist. Temporary working state may need to expire.
Read overviewShort-Term vs Long-Term Memory
Long-term memory retains useful knowledge for future work. Session state holds temporary context with configurable expiry. The two serve different purposes.
Read overview📚Technical Deep Dives
Hybrid Semantic Search
The words in a question may differ from the words in a stored memory. Semantic retrieval helps surface related information. Keyword matching supports precise terms.
Read overviewMCP and Agent Memory
MCP gives compatible agents a way to call tools. Recall Racoon uses that connection to make memory available within an agent workflow.
Read overviewClaim Lifecycle Governance
A claim can be new, accepted, stale, superseded, conflicted, or archived. Those states describe how knowledge stands within a governed memory workflow.
Read overviewKnowledge Graph Visualization
A list shows individual memories. A graph gives another view of the relationships among stored information.
Read overview📚Deployment & Operations
Self-Hosted AI Memory
Self-hosting places the memory platform in an environment you operate. It gives your team responsibility for infrastructure, access, maintenance, and recovery.
Read overviewPrivate AI Infrastructure
A private deployment starts with clear boundaries. Where is memory stored? Where do models run? Which connected tools can send or retrieve information?
Read overviewRAG and Persistent Memory
Retrieval-augmented generation brings selected information into a model’s context. Persistent memory retains information so it remains available across working sessions.
Read overviewTechnical EvaluationNeed the
Need the
implementation details?
Tell us about your client, deployment, and integration requirements. We'll discuss the technical information relevant to your setup.
Discuss Your SetupDocumentationLooking for