Memory & Knowledge MCP Servers for DeepSeek Harness

MCP server for Obsidian vaults with vector search

Lightweight, injectable knowledge bundles for AI agents. Git-versioned markdown, MCP-native, lives in your filesystem.

Local-first memory governance for AI agents: shared, reviewable, auditable memory via SQLite and MCP.

Persistent memory, self-evolving playbooks, and sandboxed REPL for AI agents

Cross-project memory sharing for Claude Code and Cowork. Namespaces let projects share knowledge without polluting global scope.

Fresh, attributed facts from your sources — cached, and invalidated the moment a source changes.

🧠 Elasticsearch-powered MCP server with hierarchical memory categorization, intelligent auto-detection, and batch review capabilities

Hebbrix MCP server — long-term memory and knowledge graph for any MCP-compatible agent (Claude Desktop, Cline, Cursor, etc.)

Stratified memory for synthetic intelligences

CLI tool for parsing .kgn/.kge files, ingesting them into a PostgreSQL knowledge graph DB, and generating AI agent context packages via

Persistent AI memory using a SQLite database with a knowledge graph.

MCP server with RAG for llms.txt documentation

100% local RAG for Obsidian, Zotero, and Claude Code — LightRAG + Ollama + MCP

Durable causal memory for AI agents — semantic recall, causal lineage, and skills. Local-first.

Semantic search engine for markdown documents. MCP server with multi-provider embeddings and Milvus vector storage.

Persistent, self-evolving memory for Claude Code — local-first, browsable Obsidian vault.

MCP stdio server that manages local memory.md with dedupe/merge/cleanup.

Persistent graph memory MCP server for AI agents using LadybugDB

Governed tacit memory for AI agents: reviewed expert know-how, given only where conditions match

MCP Server for micro-app knowledge base

Temporal memory for AI with decay and reinforcement. Two-layer storage (JSONL + Markdown).