MemSearch Jev reranking
Zilliz's cross-agent Markdown memory with an optional Jev reranker that asks yes/no relevance questions about retrieved chunks.
Overview
memsearch, from Zilliz, is a persistent, cross-platform semantic memory layer for AI coding agents, backed by Markdown files as the source of truth and Milvus for search. Memories flow across Claude Code, Codex, DeepSeek Harness, OpenClaw and OpenCode, so a conversation in one agent becomes searchable context in all others. It also distils repeated workflows into reusable agent skills and runs optional background maintenance of project and user notes. Agent users install a plugin; developers can use the CLI and Python API.
Its Jev integration is an optional remote reranker: after search returns Markdown chunks, Jev answers a yes/no relevance question about each chunk against the query and results are sorted by that probability, with no local model download. The repository publishes a Chinese and English reranking evaluation. A listing notes it is a community integration rather than an official TypeSafe SDK. It is MIT licensed.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Is this memory chunk relevant to the query? | — |
How it works
- Agents capture conversations into Markdown memory
- Search retrieves candidate chunks from Milvus
- Optional Jev reranker asks a relevance question per chunk
- Chunks are sorted by probability
- Top memories are injected into the agent's context
Strengths
- Shared memory across many agents
- Optional reranking without local models
- Published bilingual evaluation
Caveats
- Reranking sends chunks to TypeSafe
- Optional feature requiring a key
Getting started
pip install memsearch or install the plugin for your agent; enable reranking per the configuration docs.At a glance
- Type
- Library
- Status
- production
- Language
- Python
- Runtime
- agent plugins, CLI and Python API with Milvus
- Has code
- yes
- Has tests
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
Coding-agent memory: an optional Jev reranker asks Noul questions about retrieved Markdown chunks and sorts them by relevance to the query, with bilingual evaluation results.
yibie/awesome-jevMarkdown memory for coding agents with an optional Jev Noul reranker and a published English/Chinese retrieval evaluation; community integration, not an official TypeSafe SDK
AnotiaWang/awesome-decision-modelsA persistent, unified memory layer for all your AI agents (e.g. Claude Code, Codex, DSH), backed by Markdown and Milvus.
logicrw/awesome-jev-projects
Frequently asked questions
What does Jev do in memsearch?
Optionally reranks retrieved memory chunks by asking whether each is relevant.
Is it required?
No, reranking is optional.
Which agents share memory?
Claude Code, Codex, DeepSeek Harness, OpenClaw and OpenCode.
Related projects
Local long-term memory for coding agents with an optional Jev reranker that orders recalled memories.
ProjectproductionTypeScriptVolcengine's context database for agents, with a Jev rerank client that scores candidate documents with jev-latest as relevance.
LibraryproductionPythonMulti-user MCP 'agents' cloud' on Cloudflare Workers with an optional search stage that reranks candidates with Workers AI Jev scores.
MCP serverbetaTypeScriptCLI and MCP server giving coding agents Jev-powered semantic search, focused reads and list picking in about a second for about a cent.
BenchmarkbetaPythonAgent skill and Python library for Claude Code and Codex that reranks candidate lists with one Jev yes/no relevance question each.
ProjectbetaPythonLocal-first MCP server that returns only Jev-checked passages from big PDFs, Word files and spreadsheets, with exact row calculations.
MCP serverbetaTypeScript