How is Glen different from Spring AI's chat memory advisors and vector stores?

Those are per-application memory you configure and run. Glen is durable, org-shared long-term memory delivered as one MCP tool, read and written by every service and agent across your organization.

Does Spring AI already support MCP?

Yes. Spring AI has MCP client support, so connecting to Glen as an MCP tool fits the framework's existing model, no separate memory stack required.

Shared memory for Spring AI

Spring AI brings the Spring programming model to AI engineering, ChatClient, advisors, tool calling, and integrations with vector stores so JVM teams can build agents and RAG without leaving their stack. It gives you chat memory advisors and a clean abstraction over many vector databases, but you still own the infrastructure decisions: which store, which embedding model, how durable knowledge gets written and shared across services. In a microservices world, that easily becomes one memory silo per service. Glen, shared memory for AI agents, gives your Spring AI applications long-term shared memory as a single MCP tool, so every service in your organization reads from and writes to one store.

Wire Glen into a Spring AI application over MCP and any ChatClient or tool-calling agent can invoke one tool that retrieves relevant long-term context and records new facts in a single round trip. Spring AI already speaks MCP, so adding Glen is a natural extension of the model you know: instead of provisioning a vector store per service and managing chat-memory advisors and embeddings yourself, you point your clients at Glen and let one shared layer handle relevance and writes.

The advantage shows up across services. Spring AI's chat memory tracks a conversation; Glen is durable, org-shared memory that outlives any conversation and crosses service boundaries. Because it is org-scoped, the memory is shared across every Spring AI application, every agent, and every deployment in your organization, so one service learns a durable fact and another reads it next time, even a service written in a different language or an agent on a different framework. Since Glen is a standard MCP server, the same memory is readable from Claude Code, Cursor, or any other MCP client, so your JVM services and your developers draw on one knowledge layer. You connect once over OAuth or an API key and the memory compounds across your fleet.

FAQ

How is Glen different from Spring AI's chat memory advisors and vector stores?: Those are per-application memory you configure and run. Glen is durable, org-shared long-term memory delivered as one MCP tool, read and written by every service and agent across your organization.
Does Spring AI already support MCP?: Yes. Spring AI has MCP client support, so connecting to Glen as an MCP tool fits the framework's existing model, no separate memory stack required.