This gives Claude a persistent memory layer with knowledge graph capabilities for storing and retrieving visual and contextual information across conversations. The source details are minimal, but the concept is clear: instead of Claude forgetting everything between sessions, you can build up a structured knowledge base that connects entities, relationships, and visual context. You'd reach for this when building applications that need continuity, like a personal assistant that remembers your preferences and past interactions, or a research tool that builds connections between concepts over time. It's essentially turning Claude's stateless nature into something more like long-term memory with graph-based relationships to make that memory queryable and meaningful.
MEMORY_FILE_PATHPath to the memory.json file for knowledge graph storage
ALLOWED_DIRECTORIESComma-separated list of directories the server can access, or JSON array format
io.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage
io.github.mikerawsonnz/llm-orchestration-agent
io.github.mikerawsonnz/authenticated-llm-agent
labforgedev/copilot-memory-mcp
csoai-org/agent-prompt-injection-firewall-mcp
io.github.mikerawsonnz/authenticated-multi-llm-agent