Agentmemory MCP

A paid remote service that gives Claude and other AI coding agents persistent memory across sessions. It captures decisions, preferences, errors, and deployment notes from agent runs, then uses hybrid retrieval (vector plus keyword plus recency) to brief the next session with relevant context. You get 51+ memory actions over MCP, a review queue to vet memories before they become team knowledge, four scoping levels to separate operator notes from shared repo facts, and export controls. Useful when your team runs multiple agent sessions across the same codebase and you're tired of re-explaining the same migration gotchas, API quirks, or test traps every time. Ships with analytics, NOWPayments checkout, and hosted infrastructure so you skip the self-hosting work.