# zilliztech/memsearch

A persistent, unified memory layer for all your AI agents (e.g. Claude Code, Codex, DSH), backed by Markdown and Milvus.

Repository: https://github.com/zilliztech/memsearch
Canonical: https://ross.abutalabs.com/products/memsearch
Homepage: https://zilliztech.github.io/memsearch/
Language: Python
License: MIT
License Family: permissive
Topics: agent-memory, claude-code, claude-code-plugin, memory, openclaw, rag, agent, embeddings, milvus, semantic-search, ai-agents, hybrid-search, long-term-memory, opencode, skills, codex, deepseek, deepseek-harness, dsh, dsh-plugin
Last push: 2026-08-23T04:54:01+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 99, longevity 14
- inputs: {"age_days": 206, "days_push": 10, "days_rel": 11, "gap_med": 4.5, "n_releases_24m": 31}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2515, forks 236 (observed 2026-08-28T04:06:57.650986+00:00)

## What it is
memsearch is a persistent, cross-platform memory layer for AI coding agents like Claude Code, Codex, DeepSeek Harness, OpenClaw, and OpenCode. It captures agent conversations, stores them as Markdown indexed in Milvus, and recalls relevant context via hybrid semantic search across sessions and agents.

## Use cases
- give my AI coding agent persistent memory across sessions
- resume a debugging thread from a previous Claude Code session
- find why we chose a certain architecture or library last month
- share project memory between Claude Code and Codex
- search past agent conversations semantically
- do code archaeology on when and why a module changed

## When to choose
- you use Claude Code, Codex, DSH, OpenClaw, or OpenCode and lose context between sessions
- you want zero-config automatic memory capture and recall via a plugin
- you want a unified memory layer shared across multiple AI coding agents
- you prefer human-readable Markdown storage backed by Milvus hybrid search

## When to avoid
- you don't want a Milvus dependency or external vector database
- you need memory for chat agents outside coding workflows
- you need a fully self-contained memory with no indexing infrastructure
- your agent platform has no memsearch plugin

## Facets
- artifact type: plugin
- maturity: active
- function: rag, search-engine, vector-database, agent-framework, mcp
- domain: large-language-models, developer-tools
- platform: python, cli, cross-platform
- tags: agent-memory, claude-code, codex, milvus, semantic-search, hybrid-search, long-term-memory, markdown-memory, ai-agents, retrieval-augmented-generation

## Member repositories
- zilliztech/memsearch (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:57.650986+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:26:12.545117+00:00, confidence not recorded.
  - readme: https://github.com/zilliztech/memsearch (fetched 2026-08-28T04:06:57.650986+00:00, sha 174a620c2670)
  - homepage: https://zilliztech.github.io/memsearch/ (fetched 2026-08-29T10:08:31.074690+00:00, sha 1de237afb01b)
- Data as of 2026-08-30T08:39:29.467469+00:00.