"A simple memory MCP server with read/write tools and vector database search" matching MCP servers:

Gives multiple AI coding agents a shared, persistent memory by indexing your code and documentation into a vector database with AST-aware chunking and hybrid semantic/keyword search. Agents can search across projects and store, update, or forget decisions, conventions and lessons learned, all self-hosted with local embeddings.

Enables AI agents to format SQL, explain queries in plain English, analyze schemas, build queries from natural language, and generate migrations, all without requiring a database connection.

Combines Neo4j graph database with vector search using OpenAI embeddings for intelligent semantic search across knowledge graphs.

Enables Google search with advanced options (date, language, country, safe search) and webpage content extraction via the Model Context Protocol.

Provides tools for Chinese character stroke order animation with tracing practice and real-time weather queries for major cities.

Enables AI agents to maintain persistent, searchable two-layer memory with 37 tools, hybrid search, knowledge graphs, and enterprise features like authentication and backups.

A read-only MCP server for Hyperliquid that provides public market data (prices, order books, funding) and any wallet's positions, orders, and fills via MCP tools, without requiring a private key.

Enables read-only research on Reddit via the Reddit API, including finding subreddits, searching, and reading comment trees, with support for multiple configured accounts.

Enables authorized users to manage Google Ads accounts through natural language, with guardrails like role tiers, account allowlists, spend limits, and two-step confirmations to prevent accidental changes.

Enables AI agents to semantically search GitHub repository documentation by automatically fetching, vectorizing, and indexing content into an Upstash Vector database. It provides a standard MCP interface for agents to retrieve relevant documentation snippets through natural language queries.

Enables read-only access to a single Gmail mailbox, listing emails within a date range and retrieving full headers and body text of individual emails, with deep-link jump URLs for each result.

An MCP server for integrating with the Brave Search API, and it supports HTTP proxying.

Read-only MySQL/MariaDB MCP server for running SELECT queries safely, with automatic read-only enforcement and query limits.

MCP server for Microsoft SQL Server that lets LLMs explore schema and execute read-only SELECT queries, with optional stored procedure execution when explicitly enabled.

An MCP server that enables safe, audited mutation of Google Ads campaigns—creating ads, ad groups, keywords, and assets or adjusting budgets and statuses—with a dry-run default and an optional guarded remove operation, plus read-only Keyword Planner ideas.

Read-only MySQL MCP server that lets AI agents list tables, describe schemas, and run SELECT/SHOW/EXPLAIN queries with a row cap, bound to a single database for safety.

Gives AI agents a persistent vector memory by embedding text with Ollama and storing it in Qdrant, so they can save, batch-save, semantically search with payload filters, and delete memories across collections.

An MCP server for reading Twitter/X engagement data and metrics, enabling AI agents to measure tweet performance, track mentions, analyze replies, and search tweets with engagement metrics.

Enables SQL database interaction with true server-side result paging, runtime-managed connection profiles, and SSH-bridged access for PostgreSQL.

A secure MCP server for Grafana that provides read-only access by default, with a separate write-capable command for mutations.