A persistent, unified memory layer for AI coding agents, backed by Markdown and Milvus.
memsearch is easy to set up with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view zilliztech/memsearch --webOpen the official README and confirm the supported install method.
Add the server entry to your MCP client config.
Restart your agent and verify that the server tools appear.
Repository setup guidance
Strong trust signals; still review the README and permissions before production use.
Last commit was about 101 days ago.
1920 GitHub stars indicate community interest.
216 open issues signal maintenance load.
MIT license detected.
3 security/trust notes recorded.
Setup difficulty is 2/5.
Connect any AI model to 1200+ integrations (MCP, CLI, API)
An AI Gateway, registry, and proxy that sits in front of any MCP, A2A, or REST/gRPC APIs, exposing a unified endpoint with centralized discovery, guardrails and management. Optimizes Agent & Tool calling, and supports plugins.
Official Microsoft Learn MCP server and CLI tool providing real-time, trusted Microsoft docs and code samples to AI agents.
Open-source credential gateway with a built-in vault. give your AI agents access to services without exposing keys.
Memsearch gives your AI coding agents a shared memory that persists across sessions and different tools. It stores memories as simple Markdown files, making them easy to read and edit, while using Milvus for fast semantic search. This means conversations and context from one agent (like Claude Code) are automatically available to others (like Codex CLI) without extra setup.
Memsearch is a cross-platform semantic memory system designed for AI coding agents. It provides a persistent, unified memory layer that works across multiple agent platforms including Claude Code, OpenClaw, OpenCode, and Codex CLI. The system uses Markdown files as the source of truth, making memories human-readable, editable, and version-controllable. Milvus serves as a shadow index for fast hybrid search (dense vector + BM25 sparse + RRF reranking). Key features include progressive retrieval with three-layer recall (search, expand, transcript), smart deduplication via SHA-256 content hashing, and live file watching for real-time indexing. Memsearch is ideal for agent users who want plug-and-play persistent memory, and for agent developers who need a robust memory and harness engineering toolkit via CLI and Python API.
1,920
Stars
178
Forks
216
Issues
MIT
License
A persistent, unified memory layer for AI coding agents, backed by Markdown and Milvus.
AI agent users who want persistent memory without complex setup., Agent developers building custom coding assistants with memory capabilities., Teams using multiple AI coding tools who need a unified memory layer.
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
Memsearch stores conversation data in Markdown files; ensure sensitive information is not inadvertently persisted. The Milvus index is a derived cache; if deleted, it can be rebuilt from the Markdown files, but may cause temporary search unavailability.
Similar or complementary options to evaluate include metorial, mcp-context-forge, Microsoft Learn MCP Server.
Aug 20, 2026