"Finding the best memory MCPs for in-session management" matching MCP servers:

Enables AI agent sessions to reach a database through MCP with per-session, human-approved access grants, exclusive leased pool credentials, and revocable, auditable operations. It can be deployed locally or remotely with passkey-verified session administration.

Manages Supabase projects, databases, edge functions, secrets, and branches across organizations using the Supabase Management API with 24 tools.

Persistent knowledge graph MCP server with SQLite backend, offering graph traversal, fuzzy search, and temporal queries for entities and relations. It serves as a drop-in replacement for the npm mcp-server-memory with enhanced capabilities.

A drop-in replacement for Anthropic's memory server that utilizes SQLite to ensure data integrity and concurrent access. It enhances the original functionality with semantic search capabilities using vector embeddings and ONNX models.

A modular monolith MCP server connecting multiple providers (Redash, PostgreSQL, GitHub) with unified permission, security, and auditing, enabling secure data access and tool execution.

Provides persistent memory management with SQLite, enabling storage and retrieval of conversational context with metadata via MCP tools and REST API.

Provides specialized tools for portfolio health analysis, rebalance simulations, and trade execution via a single interface powered by MongoDB. It enables AI agents to manage investment portfolios while adhering to organizational governing rules, clearance guards, and rate limits.

Enables AI assistants to have persistent, self-hosted memory across sessions and devices, storing context in a SQLite database you control.

An MCP server that stores project memory in Supabase, allowing AI tools to persist context, notes, and decisions across sessions. Enables Claude Code, Codex, and similar assistants to maintain long-term project continuity through persistent memory storage.

MCP server that connects AI assistants to Redash, enabling listing of assets, read-only query execution, dashboard inspection, and alert management through the Redash API.

Exposes a governed dbt MetricFlow semantic layer to AI agents through three tools — list metrics, list dimensions and query metrics — so questions about federal contract spend are answered from defined metrics rather than raw tables. Guardrails reject undefined metrics, raw columns, SQL in filters and bulk exports, and no tool can execute SQL.

Local-first, auditable memory for AI agents. Provides durable context for MCP hosts with SQLite storage, CLI, and MCP tools for memory management.

Enables role-based order management through MCP tools, allowing users to view and refund orders according to their permissions (USER, MANAGER, ADMIN) with audit logging and JWT authentication handled by a FastAPI client.

An MCP server that provides a shared memory layer for Claude across claude.ai, Claude Code CLI, and Desktop, storing revisioned memory entries, session timelines, handoffs, and uploaded artifacts in SQLite.

Read-only MCP server that proxies The Game Pensieve API over Streamable HTTP, letting AI assistants query a personal game collection via tools like search, filters, and summaries. Supports OAuth 2.1 enforcement and owner-scoped access.

Enables coding agents to query and modify a SQLite database through twelve SQL-based tools, with safety enforced server-side via statement policies, protected tables, result caps, and snapshots. It also provides an append-only, hash-chained memory log with checkpoints so agents can record events and resume context across sessions.

Enables Claude Code and Gemini CLI to save, search, retrieve, and browse recent voice-derived notes and tags stored in SQLite over stdio, while deliberately excluding deletion and arbitrary file reads.

Enables project-scoped agent memory backed by a per-project SQLite index and raw data store, plus a global registry for discovering local memory workspaces. Agents can search summaries, retrieve original records on demand, save new memories, and link related records without sharing memory across projects.

An MCP server for managing agent memory using provenance tracking, decay-weighted retrieval, and feedback loops to optimize information recall. It allows agents to store insights in a local SQLite database and rank them based on confidence, age, and usefulness.

A simple memory system that allows LLMs to store and retrieve information via the Model Context Protocol.

A knowledge graph memory server using SQLite to provide persistent, isolated contexts for organizing information into searchable categories like work and personal projects. It features unique ID-based operations and a token-efficient serialization format designed to optimize interactions with LLMs.

Free local memory for Claude and Cursor: remember, recall and forget tools backed by a single SQLite file. Zero dependencies, MIT licensed. By Jipbap Kitchen.

Stores markdown knowledge notes in SQLite or Postgres and exposes them to AI agents over MCP, letting agents read individual sections, search, and write updates transactionally. Each write is validated, and anything that introduces broken links, missing summaries, or duplicate text is rolled back with a reason.

A Model Context Protocol server that allows users to store, retrieve, update, and delete memories using SQLite storage.

Enables coding agents to save facts, decisions and conventions into a single SQLite file and search them again in later sessions. Recall is ranked by BM25 relevance boosted by recency and scoped per project, with zero native dependencies.

A custom Memory MCP Server that acts as a cache for Infrastructure-as-Code information, allowing users to store, summarize, and manage notes with a custom URI scheme and simple resource handling.

Enables storing and querying memories in a Neo4j knowledge graph with automatic entity extraction. Supports adding episodes, searching entities and relationships, and managing graph data through natural language.

Enables assistants to append immutable, dated assertions and query what was known at a historical cutoff, using namespace-scoped hybrid FTS5/BM25 plus cosine retrieval that excludes future revisions. Exposes search, history auditing, status, and optional write/forget tools over a local stdio MCP server.

Enables AI assistants to store, score, recall, pin, sleep, wake, update, delete, and export memories with importance weighting, type-based decay, emotional valence/arousal, and dormant-memory reactivation through 14 MCP tools.

An MCP server for neuroanatomically-inspired memory consolidation using Neo4j and semantic search, enabling episodic memory encoding, semantic retrieval, and graph operations.