Knowledge2026年5月12日·1 分钟阅读

remindb — Portable Memory DB for AI Agents

remindb stores agent memory in a SQLite file with search and deltas. Install via a one-line script and serve memory to MCP-capable agents.

Agent 就绪

这个资产可以被 Agent 直接读取和安装

TokRepo 同时提供通用 CLI 命令、安装契约、metadata JSON、按适配器生成的安装计划和原始内容链接,方便 Agent 判断适配度、风险和下一步动作。

Needs Confirmation · 62/100策略:需确认
Agent 入口
任意 MCP/CLI Agent
类型
Memory
安装
Script
信任
信任等级:Established
入口
curl -fsSL https://raw.githubusercontent.com/radimsem/remindb/main/install.sh | bash
通用 CLI 安装命令
npx tokrepo install 44676243-5f9f-5ae0-85c4-c8543950d448

简介

remindb 主打“轻量 + 可携带”:把记忆放进一个 .db 文件里随处拷贝/同步,Agent 通过搜索与增量 delta 读取,而不是每次全量重读。

最适合: 需要跨会话/跨设备持久记忆的长跑型 Agent

适配: Linux/macOS/Windows;提供安装脚本;基于 SQLite 存储

配置时间: 10–20 分钟

关键事实(已验证)

  • README 提供 Linux/macOS 一行安装脚本与 Windows PowerShell 安装方式。
  • README 的源码构建说明提到需要 Go 1.26+。
  • GitHub:89 stars · 3 forks;最近更新 2026-05-12(GitHub API 验证)。

正文

想把记忆层用好,最好把它当“产品能力”来设计:

  • 明确写入什么(决策、偏好、约束)以及不写什么(原始日志、密钥)。
  • 定期对变冷的记忆节点做压缩/总结,保证搜索结果更相关、更快。
  • 用 delta/游标的增量同步方式,让 Agent 低 token 成本保持“最新状态”。

README 原文节选(verbatim)

remindb logo

remindb

Agentic memory in a single SQLite file.
Stop letting your agent re-read the same notes every session.

CI Latest release License Go version


remindb architecture

Why I built this

Coding agents already have memory. CLAUDE.md, AGENTS.md, your notes folder, that growing pile of project READMEs. Stuff persists just fine.

The problem is how the agent consumes it. Every session starts by re-reading the whole pile from scratch — every Read, every Grep, scanning raw prose the agent has already processed dozens of times. Big context windows don't fix it. A 1M-token window is still paid per call, and still can't tell yesterday's stale note from today's relevant one.

Raw markdown is the wrong shape for memory. Not because it can't hold the words — it can — but because it forces the agent to pay full freight on every read.

remindb is a single SQLite file your agent treats as long-term memory. It parses your notes (Markdown, HTML, JSON, YAML, TOON) into a structured tree, hashes every node, encodes repetitive structures compactly when it saves tokens, and surfaces the whole thing through a tight MCP tool suite.

What you get

A tree the agent can index, not skim. Instead of ls-ing a folder and reading every file to orient, the agent calls MemoryTree once. Each entry is a typed node — [heading], [list], [kv], [table], [preamble], [text], [code], [embed] — with an ID, a short label, a temperature, and a token count. Think of it as ls -la for memory: one call, a scannable index, hot stuff floats up.

A real slice (from remindb inspect --tree):

[preamble] Preamble: framework, language, project (id=3kGXxidmWBp file=CLAUDE.md temp=0.50 tok=14)

### FAQ

**remindb 是服务端吗?**
答:README 描述它以一个可携带的 SQLite 文件为核心,并提供 MCP runtime 供快速查询。

**怎么快速安装?**
答:用 README 的安装脚本,然后用 `remindb --version` 验证。

**能多 Agent 共享吗?**
答:一个 `.db` 文件确实容易共享/同步,但仍需要你自己管理访问控制与隐私。
🙏

来源与感谢

Source: https://github.com/radimsem/remindb > License: MIT > GitHub stars: 89 · forks: 3

讨论

登录后参与讨论。
还没有评论,来写第一条吧。

相关资产