Memory MCP servers

Save and query agent memory in a distributed way with Membase.

Local-first, LLM-agnostic memory layer for AI assistants.

Cognitive memory for AI agents — semantic recall, knowledge graph, and contradiction detection

Persistent 5-level hierarchical memory for AI agents. SQLite-backed, lazy-loaded.

Local agent memory with provenance tracking that detects when sources change, plus session handoffs, decision records, semantic search and a knowledge graph.

Temporal codebase knowledge graph that links test failures to code changes, learns co-edit patterns, predicts regression risk and enforces edit constraints.

Cognitive memory engine with knowledge modules, circadian rhythm awareness, emotional state tracking, hybrid PostgreSQL/Qdrant search and per-user isolation.

Local-first developer knowledge graph: watches project files, extracts entities and relationships with LLMs, and offers queries, a web dashboard and a CLI.

Local-first product graph for AI coding agents.

Persistent memory for coding agents storing architecture decisions, bug root causes and conventions, with full-text search and token-budgeted context assembly.

Local-first memory for AI coding agents on SQLite/FTS5, curated and semantically de-duplicated, with data stored as plain rows you own.

Durable memory for AI agents: fact extraction, hybrid retrieval, temporal graph. SQLite/Postgres.

Store and retrieve conversation memories from multiple LLMs in MongoDB, with timestamps and LLM identification.

Portable agent memory anchored on Solana. Local SQLite + vector recall, open export.

Local-first semantic memory: hybrid search, Markdown storage, MLX or CPU embeddings, contradiction detection, time-travel history and cross-machine git sync.

Cross-session memory for AI agents with a Source Receipt for every memory, over MCP.

Graph-powered code intelligence with semantic search and knowledge graph for AI assistants

A MCP server built for developers enabling Git based project management with project and personal…

Persistent knowledge graph MCP server for neurodivergent thinking. BM25 search, no cloud LLM.

Local-first, auditable memory with scoped user/project storage in SQLite or Postgres, read-only recall by default and opt-in reviewed writeback.

Self-hosted business context: CRM, checkpoints, automation triggers, financial ledger, tasks, playbooks, news feeds, projects and semantic memory.

Persistent memory for AI agents — 98%+ retrieval recall, 99% token savings, 44 tools

Scaffold is a Retrieval-Augmented Generation (RAG) system designed to structural understanding of large codebases. It transforms your source code into a living knowledge graph, allowing for precise, context-aware interactions that go far beyond simple file retrieval.

Deterministic, offline, read-only ADR decision memory for coding agents. No model or network calls.

Neuroscience-inspired local agent memory that stores episodes, consolidates knowledge over time and builds a personal knowledge graph with semantic recall.

Human-governed memory in local, domain-separated vaults with hybrid retrieval; also reads other coding agents' output. Requires the Mnemosyne OS desktop app.

Cross-session memory for AI agents - knowledge graph, scoring, semantic search

Encrypted mail, shared memory, and persistent context for AI agents.

Local workspace memory: indexes Markdown, CSV and session logs into a vector store with hybrid search, cross-session memory and knowledge graph visualization.

Web search via embedded SearXNG, content extraction and library docs indexing with hybrid FTS5 and semantic search using built-in Qwen3 embeddings.

Structural memory for coding agents: indexes TypeScript/JavaScript repos for callers, callees, symbol locations and branch-aware blast-radius impact.

Event-sourced agent memory on a hash-chained log with an embedded Kuzu temporal knowledge graph, salience-based forgetting and review-gated consolidation.

Cerefox — user-owned shared memory for AI agents. CLI + stdio MCP server + web UI + ingestion for a knowledge base on your own Supabase project (or fully self-hosted with Cerefox Local).

Turns your task manager into agent memory with hybrid retrieval over TickTick or an Obsidian vault, without a separate vector database.

Reasoning, code, anti-deception, memory harness MCP tools. Stdio or HTTPS api.ejentum.com/mcp

Bitemporal agent memory with deterministic fact supersession, point-in-time recall, memory lineage, conflict tracking and tamper-evident audit trails.

Self-hosted persistent agent memory with entity resolution, PII detection, AES-256-GCM encryption at rest and hybrid search.

Cross-LLM communication and memory sharing so different AI models can collaborate and share context across conversations.

A local typed-edge graph of people, projects, and decisions that externalises memory.

Evidence-grounded agent memory with mandatory provenance, on local SQLite or Cloudflare D1

Memory database for LLM agents — persistent semantic + keyword memory, project namespacing, served o

Persistent memory and codebase context for coding agents: FAISS vector store, AST symbol index, knowledge graph, episodic journal and cross-agent handoff.

Local-first cross-agent memory in one SQLite file with a layered structure, Ebbinghaus-style forgetting and detection of code drifting from past decisions.

Persistent local memory for Claude Code: indexes all sessions verbatim into SQLite and ChromaDB for semantic recall across your history, with no API calls.

Local-first long-horizon agent memory: source-traceable recall, belief revision, selective forgetting and a guard against memory-driven destructive actions.

Use Mistral AI for chat, OCR, audio, vision, agents, embeddings, moderation, classification, files and batch jobs, with French/English prompts.

Open, Git-native memory protocol for MCP agents. Portable, versioned, human-editable.

Persistent, verbatim, searchable memory for AI assistants — one memory across every MCP client.