Memory & Knowledge MCP Servers for OpenAI Agents SDK

Local-first, graph-linked persistent memory for AI coding agents (MCP server + CLI)

Tropicalia MCP server for AI Agents

MCP360 Unified Gateway - Access all MCP servers through a single unified interface with search and execute capabilities

A production-ready MCP server containing multiple embedded servers (filesystem, web-search, database, calculator, system)

MCP server for graph-aware Obsidian vault traversal

MCP server for walnut structured context — read, save, and manage walnuts from any MCP client

Personal, passkey-protected memory for Claude, Codex, and other MCP clients

MCP server for enabling memory of multiple projects for through a knowledge graph

Dynamic MCP server for the Yantrix API portfolio — 30+ tools across research, memory, language, legal, security, intelligence, and more.

Persistent memory for AI agents — pgvector + BM25 keyword search with RRF fusion, credential scrubbing, auto-consolidation.

Local-first MCP server + CLI that gives terminal AI agents an endless semantic memory via Anchored Compacting, keeping the context window

Persistent agent memory with a decision layer: replay, restore, verify, or none — explainable.

MCP server for agent-safe Obsidian vault access with lint validation

Cross-agent handoff: the next coding agent starts from the last one's facts, not a blank page.

Salesforce AgentForce knowledge graph — 40 nodes, Einstein Trust Layer, 11x fewer tokens than RAG.

[DEPRECATED - use EchoVault] MCP Server for AI memory with PostgreSQL

28-layer cognitive substrate with cross-session ToT evolutionary memory and self-telemetry tools

NVIDIA NemoClaw knowledge graph — 55 nodes, F1 0.576 (+269% vs RAG), 11x fewer tokens. MCP-native.

Prior art for AI agents. init wires DID/MCP; claim --yes puts DID on commons (ask alone does not).

Review-gated declarative long-term memory with bounded read-only MCP access.

An MCP server that gives AI agents persistent, intent-based memory with vector embeddings and semantic indexing.