Skip to content
Back to search

ReMe

ReMe: Memory Management Kit for Agents - Remember Me, Refine Me.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/agentscope-ai/ReMe with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/agentscope-ai/ReMe'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/agentscope-ai--reme/badge.svg)](/listings/agentscope-ai--reme)

HTML

<a href="/listings/agentscope-ai--reme"><img src="/api/listings/agentscope-ai--reme/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Installation

ReMe requires Python 3.11+.

Install from pip:

pip install "reme-ai[core]"

Install from source:

git clone https://github.com/agentscope-ai/ReMe.git
cd ReMe
pip install -e reme_studio -e ".[core]"
cd reme_studio
npm ci
npm run build:static
cd ..

The static build requires Node.js 22.13 or newer and makes Studio available from the source tree.

Docker

With Docker and Compose 2.24.0+, build and start ReMe with the bundled Studio:

mkdir -p .reme
docker compose up --build -d

Open <http://127.0.0.1:2333>. The complete workspace persists in `./.reme`. On Linux, set `REME_UID` and `REME_GID` to your user's IDs when they differ from 1000. See [Docker deployment](https://reme.agentscope.io/en/docker) for model credentials, custom paths, published images, and upgrades.

Environment Variables

Configure environment variables when you want LLM-powered memory evolution or embedding retrieval. Embeddings are disabled by default, so the default setup does not start an embedding model or require an embedding API key.

cat > .env <<'EOF'
# Optional: used only after embedding components are explicitly enabled in the config.
# EMBEDDING_API_KEY=sk-xxx
# EMBEDDING_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1

# Required for auto_memory, auto_resource, auto_dream, and proactive refresh.
LLM_API_KEY=sk-xxx
LLM_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
EOF

Basic file operations, BM25 search, wikilink traversal, and reading proactive topics can run without LLM credentials.

> [!NOTE] > To enable embedding-based semantic retrieval, uncomment `components.as_embedding` and > `components.embedding_store` in [`reme/config/default.yaml`](reme/config/default.yaml), then change > `components.file_store.default.embedding_store` from `""` to `default`. See the > [memory search guide](https://reme.agentscope.io/en/memory_search) for details.

Start the Service

reme start

The default service address is `127.0.0.1:2333`. If the port is occupied, specify another port:

reme start service.port=8181
# reme start workspace_dir=/tmp/reme-demo service.port=8181
reme version
reme health_check
reme help
curl -s http://127.0.0.1:2333/version -H 'Content-Type: application/json' -d '{}'

5-Minute Memory Demo

With the service running, write a memory node, let ReMe index it, then retrieve it:

reme write \
  path=digest/wiki/quick-start-demo \
  name="Quick Start Demo" \
  description="A first ReMe memory node" \
  content="# Quick Start Demo

ReMe stores agent memory as readable Markdown.

Related: [[digest/wiki/memory-as-file.md]]"

reme search query="agent memory markdown" limit=5
reme read path=digest/wiki/quick-start-demo start_line=1 end_line=20

The generated file is ordinary Markdown with frontmatter:

---
name: Quick Start Demo
description: A first ReMe memory node
---

# Quick Start Demo

ReMe stores agent memory as readable Markdown.

Related: [[digest/wiki/memory-as-file.md]]

ReMe Studio (Optional)

The `core` installation includes Studio. After starting ReMe, open <http://127.0.0.1:2333/> to browse, edit, and search the workspace. To add Studio to a base installation, use `pip install "reme-ai[web]"`. See the [ReMe Studio guide](https://reme.agentscope.io/en/workspace/studio) for source builds, configuration, and development.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 474 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <p align="center">
  2. <img src="https://raw.githubusercontent.com/agentscope-ai/ReMe/main/docs/figure/reme_logo.png" alt="ReMe Logo" width="50%">
  3. </p>
  4. <p align="center">
  5. <a href="https://pypi.org/project/reme-ai/"><img src="https://img.shields.io/badge/python-3.11+-blue" alt="Python Version"></a>
  6. <a href="https://pypi.org/project/reme-ai/"><img src="https://img.shields.io/pypi/v/reme-ai.svg?logo=pypi" alt="PyPI Version"></a>
  7. <a href="https://pepy.tech/project/reme-ai/"><img src="https://img.shields.io/pypi/dm/reme-ai" alt="PyPI Downloads"></a>
  8. <a href="https://github.com/agentscope-ai/ReMe"><img src="https://img.shields.io/github/commit-activity/m/agentscope-ai/ReMe?style=flat-square" alt="GitHub commit activity"></a>
  9. <a href="./LICENSE"><img src="https://img.shields.io/badge/license-Apache--2.0-black" alt="License"></a>
  10. <a href="https://reme.agentscope.io"><img src="https://img.shields.io/badge/docs-ReMe-blue" alt="Documentation"></a>
  11. <a href="./README.md"><img src="https://img.shields.io/badge/English-Click-yellow" alt="English"></a>
  12. <a href="./README_ZH.md"><img src="https://img.shields.io/badge/简体中文-点击查看-orange" alt="简体中文"></a>
  13. <a href="https://github.com/agentscope-ai/ReMe"><img src="https://img.shields.io/github/stars/agentscope-ai/ReMe?style=social" alt="GitHub Stars"></a>
  14. <a href="https://deepwiki.com/agentscope-ai/ReMe"><img src="https://img.shields.io/badge/DeepWiki-Ask_Devin-navy.svg" alt="DeepWiki"></a>
  15. </p>
  16. <p align="center">
  17. <a href="https://trendshift.io/repositories/20528" target="_blank"><img src="https://trendshift.io/api/badge/repositories/20528" alt="agentscope-ai%2FReMe | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
  18. </p>
  19. <p align="center">
  20. <strong>A local-first, self-evolving personal knowledge base for AI agents.</strong><br>
  21. </p>
  22. > Previous versions: [0.3.x](https://github.com/agentscope-ai/ReMe/tree/reme_v3) ·
  23. > [0.2.x](https://github.com/agentscope-ai/ReMe/tree/v0.2.0.6) ·
  24. > [MemoryScope](https://github.com/agentscope-ai/ReMe/tree/memoryscope_branch)
  25. ## ✨ Why ReMe?
  26. 🧠 ReMe turns conversations and resources into readable, editable, searchable, and interconnected Markdown memory. Agents
  27. such as QwenPaw and DeepSeek Harness can share the same workspace to retrieve, maintain, and evolve knowledge, while
  28. users retain control of the durable files.
  29. - **Memory as File, File as Memory**: ReMe stores durable memory as ordinary Markdown with frontmatter and wikilinks.
  30. Users and agents can inspect, edit, move, sync, and back it up with familiar tools, while indexes and generated
  31. metadata remain rebuildable.
  32. - **Self-evolving knowledge base**: ReMe progressively turns conversations and resources into daily notes and long-term
  33. knowledge, preserving sources while refining facts, preferences, procedures, and relationships over time.
  34. - **Recall is precise and context-aware.** BM25, optional embeddings, and wikilink expansion retrieve relevant
  35. line-level passages and their relationships without loading the entire knowledge base into the agent context.
  36. - **One memory workspace works across agents.** Personal assistants, coding agents, and other agent runtimes can share
  37. the same local workspace through native integrations, SKILL.md, CLI, HTTP, MCP, or Python APIs.
  38. <p align="center">
  39. <img src="docs/figure/design-philosophy.svg" alt="ReMe Design Philosophy" width="92%">
  40. </p>
  41. ## 📰 Latest Updates
  42. <!-- Keep at most 5 updates expanded; move older entries into the collapsed section below. -->
  43. - [2026.10] - **[Decision-model context compaction](docs/en/context_compaction.md)**: score old tool calls and results
  44. against the next action, preserve protected history, and return a recoverable compaction proposal. Includes an SVG
  45. walkthrough, two live-model integration scenarios, and a 14-query fixed-summary Top-N benchmark with a forced Hermes summary baseline.
  46. - [2026.10] - **[ReMe Multimodal Memory](https://reme.agentscope.io/en/blog_20261009) update**: Auto Memory reads images
  47. with their conversation context and preserves image sources; Auto Resource turns image files into searchable daily
  48. cards linked to the originals. With Qwen3.8-Max, ReMe Image scores
  49. [72.49% on MemLens-256K](benchmark/memlens/docs/en/evaluation.md) and
  50. [60.69% on MobileMem](benchmark/mobilemem/docs/en/evaluation.md) in LLM-Judge accuracy; the reports also cover Kimi
  51. and DeepSeek.
  52. - [2026.10] - **[ReMe Studio Playground](https://reme.agentscope.io/studio/?lang=en) is live**: explore example memory
  53. files, edit Markdown, and browse linked memory graphs right in your browser—no installation or backend required.
  54. Everyone is welcome to [try it out](https://reme.agentscope.io/studio/?lang=en)!
  55. <p align="center">
  56. <a href="https://reme.agentscope.io/studio/?lang=en">
  57. <img src="https://raw.githubusercontent.com/agentscope-ai/ReMe/main/reme_studio/figures/studio-overview.png" alt="ReMe Studio workspace preview — click to try the Playground" width="480" style="margin: 0 auto;">
  58. </a>
  59. </p>
  60. - [2026.09] - **[ReMe Memory Tags](https://reme.agentscope.io/en/blog_20260920) published**: an introduction
  61. to file-native entity tags, rebuildable tag indexes, and tag-filtered memory search.
  62. - [2026.09] - **[Hermes Agent memory provider](https://reme.agentscope.io/en/integrations/hermes) available**: choose HTTP or embedded
  63. mode for automatic recall before model calls and asynchronous `auto_memory` after completed turns. The integration
  64. supports Hermes Agent 0.21+ and includes profile-aware background work.
  65. <details>
  66. <summary>More updates</summary>
  67. - [2026.09] - **[OpenClaw plugin](https://reme.agentscope.io/en/integrations/openclaw) released**: install it from
  68. [ClawHub](https://clawhub.ai/agentscope-ai/plugins/reme-openclaw-plugin) or
  69. [npm](https://www.npmjs.com/package/@agentscope-ai/reme-openclaw-plugin) to add native memory recall, automatic
  70. conversation capture, and scheduled consolidation to OpenClaw.
  71. - [2026.09] - **[DeepSeek Harness plugin](https://reme.agentscope.io/en/integrations/dsh) released**: install it from
  72. [Awesome DSH Plugin](https://awesome-dsh-plugin.com/p/agentscope-ai/ReMe--integrations-dsh/) or
  73. [npm](https://www.npmjs.com/package/@agentscope-ai/reme-dsh-plugin) for long-term-memory guidance, `reme_search`,
  74. automatic memory, Auto Dream, and ReMe Status.
  75. - [2026.08] - **ReMe blog published**: the [ReMe blog](https://reme.agentscope.io/en/reme-blog) introduces the
  76. local-first memory architecture, self-evolving workflows, hybrid search, proactive discovery, and benchmark results.
  77. - [2026.08] - **New ReMe ecosystem plugins**: [Daily Paper](https://reme.agentscope.io/en/plugins/daily-paper)
  78. discovers and analyzes papers and generates file-native briefs, while
  79. [Auto Fin](https://reme.agentscope.io/en/plugins/auto-fin) researches the latest 24 hours of topic-related CLS news
  80. and builds traceable reports with local memory. Try them out.
  81. - [2026.08] - **Plugin development support released**: use [Plugin Development](https://reme.agentscope.io/en/plugin_development) and
  82. [Plugin Management](https://reme.agentscope.io/en/plugin_management) to extend ReMe with Components, Steps, and Jobs. Contributions and
  83. new community plugins are welcome.
  84. - [2026.08] - ReMe's [experience-driven enhancement method](https://reme.agentscope.io/en/benchmarks/toolmemory) for
  85. agent tool use is available on [arXiv:2608.03403](https://arxiv.org/abs/2608.03403).
  86. - [2026.07] - Our
  87. paper [Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution](https://aclanthology.org/2026.findings-acl.829/)
  88. has been accepted to Findings of ACL 2026.
  89. </details>
  90. ## 🚀 Quick Start
  91. ### Installation
  92. ReMe requires Python 3.11+.
  93. Install from pip:
  94. ```bash
  95. pip install "reme-ai[core]"
  96. ```
  97. Install from source:
  98. ```bash
  99. git clone https://github.com/agentscope-ai/ReMe.git
  100. cd ReMe
  101. pip install -e reme_studio -e ".[core]"
  102. cd reme_studio
  103. npm ci
  104. npm run build:static
  105. cd ..
  106. ```
  107. The static build requires Node.js 22.13 or newer and makes Studio available from the source tree.
  108. Optional decision models use `pip install "reme-ai[core,decision]"`. Configure a TypeSafe API key and run the
  109. service demo as described in [Decision model configuration](https://reme.agentscope.io/en/configuration#decision-model-configuration).
  110. ### Docker
  111. With Docker and Compose 2.24.0+, build and start ReMe with the bundled Studio:
  112. ```bash
  113. mkdir -p .reme
  114. docker compose up --build -d
  115. ```
  116. Open <http://127.0.0.1:2333>. The complete workspace persists in `./.reme`. On Linux, set `REME_UID` and `REME_GID` to your
  117. user's IDs when they differ from 1000. See [Docker deployment](https://reme.agentscope.io/en/docker) for model credentials,
  118. custom paths, published images, and upgrades.
  119. ### Environment Variables
  120. Configure environment variables when you want LLM-powered memory evolution or embedding retrieval. Embeddings are
  121. disabled by default, so the default setup does not start an embedding model or require an embedding API key.
  122. ```bash
  123. cat > .env <<'EOF'
  124. # Optional: used only after embedding components are explicitly enabled in the config.
  125. # EMBEDDING_API_KEY=sk-xxx
  126. # EMBEDDING_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
  127. # Required for auto_memory, auto_resource, auto_dream, and proactive refresh.
  128. LLM_API_KEY=sk-xxx
  129. LLM_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
  130. EOF
  131. ```
  132. Basic file operations, BM25 search, wikilink traversal, and reading proactive topics can run without LLM credentials.
  133. > [!NOTE]
  134. > To enable embedding-based semantic retrieval, uncomment `components.as_embedding` and
  135. > `components.embedding_store` in [`reme/config/default.yaml`](reme/config/default.yaml), then change
  136. > `components.file_store.default.embedding_store` from `""` to `default`. See the
  137. > [memory search guide](https://reme.agentscope.io/en/memory_search) for details.
  138. ### Start the Service
  139. Start ReMe in one terminal and leave it running:
  140. ```bash
  141. reme start
  142. ```
  143. The default service address is `127.0.0.1:2333`. If the port is occupied, specify another port:
  144. ```bash
  145. reme start service.port=8181
  146. # reme start workspace_dir=/tmp/reme-demo service.port=8181
  147. ```
  148. From another terminal, call the running service. The CLI discovers its configuration; the `curl` example uses the
  149. default port `2333`:
  150. ```bash
  151. reme version
  152. reme health_check
  153. reme help
  154. curl -s http://127.0.0.1:2333/version -H 'Content-Type: application/json' -d '{}'
  155. ```
  156. ### 5-Minute Memory Demo
  157. With the service running, write a memory node, let ReMe index it, then retrieve it:
  158. ```bash
  159. reme write \
  160. path=digest/wiki/quick-start-demo \
  161. name="Quick Start Demo" \
  162. description="A first ReMe memory node" \
  163. content="# Quick Start Demo
  164. ReMe stores agent memory as readable Markdown.
  165. Related: [[digest/wiki/memory-as-file.md]]"
  166. reme search query="agent memory markdown" limit=5
  167. reme read path=digest/wiki/quick-start-demo start_line=1 end_line=20
  168. ```
  169. The generated file is ordinary Markdown with frontmatter:
  170. ```markdown
  171. ---
  172. name: Quick Start Demo
  173. description: A first ReMe memory node
  174. ---
  175. # Quick Start Demo
  176. ReMe stores agent memory as readable Markdown.
  177. Related: [[digest/wiki/memory-as-file.md]]
  178. ```
  179. ### ReMe Studio (Optional)
  180. The `core` installation includes Studio. After starting ReMe, open <http://127.0.0.1:2333/> to browse, edit, and search
  181. the workspace. To add Studio to a base installation, use `pip install "reme-ai[web]"`. See the
  182. [ReMe Studio guide](https://reme.agentscope.io/en/workspace/studio) for source builds, configuration, and development.
  183. ## 🤝 Use ReMe with Your Agent
  184. ReMe can run as a local memory service accessed through the CLI, HTTP API, or MCP server, or it can be embedded in the
  185. host process through its Python API. Host integrations can add memory guidance, recall, and capture to the agent
  186. lifecycle according to the capabilities of each runtime.
  187. | Agent | Recommended path | Available after integration |
  188. | ------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- |
  189. | **DeepSeek Harness** | Install [`@agentscope-ai/reme-dsh-plugin`](https://reme.agentscope.io/en/integrations/dsh) with `dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin`. | Configurable memory guidance, `reme_search`, automatic turn capture, scheduled Auto Dream, and ReMe Status. |
  190. | **OpenClaw** | Install [`@agentscope-ai/reme-openclaw-plugin`](https://reme.agentscope.io/en/integrations/openclaw) with `openclaw plugins install clawhub:@agentscope-ai/reme-openclaw-plugin`. | Native memory tools, recall before user-triggered runs, and automatic turn capture. |
  191. | **QwenPaw** | Embed ReMe in-process through its Python API. | Reuse the host lifecycle and model config while keeping memory local and file-based. |
  192. | **Claude Code** | Start ReMe, then install the [Claude Code plugin](integrations/claude_code/README.md). | Automatic recall, batched completed-turn recording, and MCP tools. |
  193. | **Qoder** | Start ReMe, then install the [Qoder plugin](integrations/qoder/README.md). | Automatic recall, retryable completed-turn recording, MCP tools, and the bundled skill. |
  194. | **Hermes** | Install [the ReMe provider](https://reme.agentscope.io/en/integrations/hermes) and choose HTTP or embedded mode. | Recall before model calls and asynchronous `auto_memory` after each completed turn. |
  195. | **Codex** | Start ReMe, then install the [Codex plugin](integrations/codex/README.md). | Native hooks for automatic recall and batched recording, plus MCP tools. |
  196. | **Other CLI agents** | Install or copy the [ReMe Memory skill](skills/reme_memory/SKILL.md). | Search, read, and write memory through the CLI. |
  197. <p align="center"><b>Integration demos</b></p>
  198. <table>
  199. <tr>
  200. <td align="center"></td>
  201. <td width="45%" align="center"><b>Auto Memory</b></td>
  202. <td width="45%" align="center"><b>Auto Dream</b></td>
  203. </tr>
  204. <tr>
  205. <td align="center"><b>QwenPaw</b></td>
  206. <td width="45%">
  207. <img src="docs/figure/qwenpaw-auto-memory.gif" alt="QwenPaw Auto Memory demo" width="100%">
  208. </td>
  209. <td width="45%">
  210. <img src="docs/figure/qwenpaw-auto-dream.gif" alt="QwenPaw Auto Dream demo" width="100%">
  211. </td>
  212. </tr>
  213. <tr>
  214. <td align="center"><b>Claude Code</b></td>
  215. <td width="45%">
  216. <img src="docs/figure/cc-auto-memory.gif" alt="Claude Code Auto Memory demo" width="100%">
  217. </td>
  218. <td width="45%">
  219. <img src="docs/figure/cc-auto-dream.gif" alt="Claude Code Auto Dream demo" width="100%">
  220. </td>
  221. </tr>
  222. </table>
  223. ## 🧠 How ReMe Works
  224. > Memory as File, File as Memory.
  225. ReMe treats **memory as files**, progressively processing filtered conversation source records and external resources
  226. from `session/` and `resource/` into `daily/`, then `digest/`. The default workspace is `.reme/` under the current
  227. directory; `workspace_dir=...` selects a different user-owned location.
  228. ### Workspace Layout
  229. ```text
  230. <workspace_dir>/
  231. ├── metadata/ # Rebuildable indexes, graphs, catalogs, and caches
  232. ├── session/ # Conversation source records and agent sessions
  233. │ ├── dialog/
  234. │ │ └── <session_id>.jsonl # Source messages saved by auto_memory
  235. │ └── claude_code/
  236. │ └── <session_id>.jsonl # ReMe copy used by auto_memory_cc
  237. ├── mem_session/ # Generated agent-wrapper sessions/config, not user memory
  238. │ ├── agentscope/
  239. │ ├── claude_config/
  240. │ └── codex/
  241. ├── resource/ # External raw materials
  242. │ ├── <resource>.<ext> # Root-level files enter today's daily layer
  243. │ └── YYYY-MM-DD/
  244. │ └── <resource>.<ext>
  245. ├── daily/ # Lightly processed memory: daily facts, conversation summaries, resource readings
  246. │ ├── YYYY-MM-DD.md
  247. │ └── YYYY-MM-DD/
  248. │ ├── <generated_name>.md # Topic-named conversation or resource card
  249. │ └── interests.yaml
  250. └── digest/ # Long-term memory: personal facts, procedural experience, knowledge nodes
  251. ├── personal/
  252. │ └── {topic/event}.md
  253. ├── procedure/
  254. │ └── {topic/event}.md
  255. └── wiki/
  256. └── {topic/event}.md
  257. ```
  258. <p align="center">
  259. <img src="docs/figure/reme-overview.svg" alt="ReMe file-based memory system overview" width="92%">
  260. </p>
  261. ### Memory Lifecycle
  262. ReMe follows a capture → index → consolidate → recall loop. Workspace files remain the durable source of truth;
  263. everything under `metadata/` is rebuildable.
  264. | Capability | Entry point | What it does | Output |
  265. | ------------------------------------------- | ----------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------- |
  266. | [`auto_memory`](https://reme.agentscope.io/en/auto_memory) | Agent hook or `reme auto_memory` | Distills useful conversation facts while preserving a filtered conversation source record. | `session/dialog/*.jsonl`, `daily/<date>/<generated-name>.md` |
  267. | [`auto_resource`](https://reme.agentscope.io/en/auto_resource) | Resource watcher or `reme auto_resource` | Turns files under `resource/` into source-linked, content-named daily cards. | `daily/<date>/<resource-card>.md` |
  268. | [`auto_index`](https://reme.agentscope.io/en/memory_search) | Background watcher or `reme reindex` | The watcher ingests Markdown from `daily/` and `digest/`; `reindex` only rebuilds BM25 and embeddings from already-ingested chunks. | Searchable chunks, BM25, wikilink graph, and optional vectors |
  269. | [`auto_dream`](https://reme.agentscope.io/en/auto_dream) | `dream_cron` or `reme auto_dream` | By default, extracts up to five reusable units from changed files in the latest two-day window, then creates, corroborates, refines, or corrects digest nodes. | `digest/**` |
  270. | [`proactive_read`](https://reme.agentscope.io/en/proactive) | `reme proactive_read` before an agent decides to act | Reads topics generated by the independent proactive refresh flow; the host agent decides whether and how to mention them. | Structured topics from `daily/<date>/interests.yaml` |
  271. <table>
  272. <tr>
  273. <td align="center" width="50%">
  274. <img src="docs/figure/memory-as-file.svg" alt="Memory as File" width="92%">
  275. </td>
  276. <td align="center" width="50%">
  277. <img src="docs/figure/auto-memory-resource.svg" alt="Auto Memory and Resource" width="92%">
  278. </td>
  279. </tr>
  280. <tr>
  281. <td align="center" width="50%">
  282. <img src="docs/figure/auto-dream-and-proactive.svg" alt="Auto Dream and Proactive" width="92%">
  283. </td>
  284. <td align="center" width="50%">
  285. <img src="docs/figure/auto-index-and-memory-search.svg" alt="Auto Index and Memory Search" width="92%">
  286. </td>
  287. </tr>
  288. </table>
  289. Search returns matching chunks with line ranges and bounded wikilink neighbors. Optional vector results are fused with
  290. BM25 through reciprocal rank fusion (RRF).
  291. > [!IMPORTANT]
  292. >
  293. > `proactive_read` only reads and exposes interest topics produced by proactive refresh. It does not independently browse the web,
  294. > send notifications, or rewrite the knowledge base; the host agent decides whether and how to act on a topic.
  295. ## 📊 Benchmarks
  296. ReMe evaluates multi-session and long-context text and multimodal memory with agentic search-and-read workflows. The figures below are the
  297. published reference runs in this repository; model, prompt, dataset, and judging details are documented with each
  298. benchmark.
  299. | Benchmark | Setting | Sample size | Agentic score | Focus |
  300. | --------------------------------------------------------------------------- | ------------ | -----------------------: | ------------: | ------------------------------------------------------------------ |
  301. | **[LongMemEval cleaned-s](https://reme.agentscope.io/en/benchmarks/longmemeval)** | **Overall** | **500 questions** | **89.4%** | Cross-session retrieval, knowledge updates, and temporal reasoning |
  302. | [BEAM](https://reme.agentscope.io/en/benchmarks/beam) | 100K context | 20 cases / 400 questions | 66.1% | Ten types of long-context memory tasks |
  303. | [BEAM](https://reme.agentscope.io/en/benchmarks/beam) | 1M context | 35 cases / 700 questions | 65.0% | Ultra-long conversation settings |
  304. | [MemLens-256K](benchmark/memlens/docs/en/evaluation.md) | ReMe Image · Qwen3.8-Max | 189 / 195 questions | 72.49% | Multimodal long-history question answering |
  305. | [MobileMem-Omni](benchmark/mobilemem/docs/en/evaluation.md) | ReMe Image · Qwen3.8-Max | 20 users / 1,333 questions | 60.69% | Personal visual memory and reasoning |
  306. ReMe also achieved a **0.580 PROC score across five user personas** in the repository's
  307. [π-Bench evaluation](https://reme.agentscope.io/en/benchmarks/pibench), 2.4% above NanoBot under the same test-model configuration. PROC
  308. measures proactive handling of hidden intent, clarification, cross-session preferences and conventions, task
  309. dependencies, and underspecified requests.
  310. ## 🧩 Extensions and Plugins
  311. Plugins are optional Python distributions that contribute Component, Step, or Job backends and configuration. They are
  312. installed separately and enabled explicitly by configuration. Daily Paper and Auto Fin are independently packaged
  313. plugins; see their documentation for [Daily Paper](https://reme.agentscope.io/en/plugins/daily-paper) and
  314. [Auto Fin](https://reme.agentscope.io/en/plugins/auto-fin).
  315. | Plugin | Capability |
  316. | ------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------- |
  317. | [Daily Paper](https://reme.agentscope.io/en/plugins/daily-paper) | Discover and rank papers, analyze PDFs with an agent, and generate file-native notes and a five-minute brief. |
  318. | [Auto Fin](https://reme.agentscope.io/en/plugins/auto-fin) | Fetch topic-related CLS news, search ReMe history, and generate wikilink-backed Markdown reports. |
  319. See [Plugin Management](https://reme.agentscope.io/en/plugin_management) to install, inspect, validate, enable, and uninstall ReMe plugins.
  320. ## 📚 Documentation
  321. These guides cover the main user workflows and the runtime contracts implemented by the current code.
  322. | Guide | What you will learn |
  323. | ------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- |
  324. | [Quick Start](https://reme.agentscope.io/en/quick_start) | Install ReMe, start the service, and run the first file and memory operations. |
  325. | [Configuration](https://reme.agentscope.io/en/configuration) | Configure the workspace, models, Service, Jobs, Components, plugins, and CLI overrides. |
  326. | [Services and Deployment](https://reme.agentscope.io/en/services) | Use HTTP, SSE, MCP, and Studio while respecting the default security boundary. |
  327. | [Memory as File](https://reme.agentscope.io/en/memory_as_file) | Understand workspace layers, frontmatter, wikilinks, chunks, and the file-as-source-of-truth model. |
  328. | [Auto Memory](https://reme.agentscope.io/en/auto_memory) | Preserve source conversations and distill reusable daily memory cards. |
  329. | [Auto Resource](https://reme.agentscope.io/en/auto_resource) | Import supported text and image resources as source-linked daily cards. |
  330. | [Auto Dream](https://reme.agentscope.io/en/auto_dream) and [Auto Link](https://reme.agentscope.io/en/auto_link) | Consolidate daily notes into evolving digest nodes and readable wikilink relationships. |
  331. | [Memory Search](https://reme.agentscope.io/en/memory_search) | Use BM25, optional vectors, RRF fusion, line-range recall, and progressive link expansion. |
  332. | [Proactive](https://reme.agentscope.io/en/proactive) | Read interest topics safely and integrate them into a host agent's decision flow. |
  333. | [Application Scenarios](https://reme.agentscope.io/en/reme_scene) | Follow concrete financial research, coding-memory, and personal knowledge-base examples. |
  334. | [Framework](https://reme.agentscope.io/en/framework) | Understand Application, Job, Step, Component, service, configuration, and lifecycle boundaries. |
  335. | [Agent Integrations](https://reme.agentscope.io/en/integrations) | Choose an interface and connect DSH, Claude Code, Codex, Qoder, OpenClaw, Hermes, or another agent. |
  336. | [DSH plugin](https://reme.agentscope.io/en/integrations/dsh), [Claude Code plugin](https://reme.agentscope.io/en/integrations/claude-code), and [Qoder plugin](https://reme.agentscope.io/en/integrations/qoder) | Configure host-native recall, automatic capture, consolidation, and diagnostics. |
  337. | [CLI and Job API](https://reme.agentscope.io/en/reference/cli) | Learn command syntax and use the generated default Job parameter reference. |
  338. | [Operations and Recovery](https://reme.agentscope.io/en/operations) | Diagnose services, maintain indexes, and back up, migrate, or recover a workspace. |
  339. | [ReMe Blog](https://reme.agentscope.io/en/reme-blog) | Read the product story, design rationale, examples, and benchmark summary. |
  340. ## 🛠️ Common Commands
  341. Run `reme help` for the full job list. Common workspace and maintenance commands are:
  342. | Command | Purpose |
  343. | ----------------------------------------- | --------------------------------------------------------------------------------- |
  344. | `reme status` | Show stateful data-component memory estimates and process RSS. |
  345. | [`reme search`](https://reme.agentscope.io/en/memory_search) | Retrieve memory with BM25 and wikilinks by default, plus vectors when enabled. |
  346. | `reme read` / `reme write` / `reme edit` | Inspect and maintain Markdown memory files. |
  347. | `reme traverse` / `reme graph_snapshot` | Explore wikilink neighborhoods or the category-rooted digest graph. |
  348. | `reme chat` | Stream a read-only, workspace-aware agent conversation. Requires LLM credentials. |
  349. | `reme reindex` | Rebuild BM25 and embedding indexes from already-ingested chunks. |
  350. ## 🤝 Community and Contributing
  351. - **Issues, requests, and help**: Check [Open Issues](https://github.com/agentscope-ai/ReMe/issues) first. If there is no
  352. related discussion, open one with the background, expected behavior, and impact scope.
  353. - **Code contributions**: Before making changes, read the repository's
  354. [contribution guide](https://reme.agentscope.io/en/contributing). Source, schemas, and tests are the authoritative architecture and
  355. extension guide.
  356. - **Documentation contributions**: Update the canonical files under `docs/en/`, `docs/zh/`, or the relevant package
  357. directory in this repository. The documentation site is generated from these files.
  358. - **Commit convention**: Conventional Commits are recommended, for example `feat(search): add link expansion option` or
  359. `docs(zh): update quick start`.
  360. - **Pre-submit checks**: Before submitting a PR, try to run `pre-commit run --all-files` and `pytest`. If tests that
  361. depend on LLMs, embeddings, or external services cannot run, explain that in the PR.
  362. - **Documentation**: Visit [reme.agentscope.io](https://reme.agentscope.io).
  363. ### Contributors
  364. Thanks to everyone who has contributed to ReMe:
  365. <a href="https://github.com/agentscope-ai/ReMe/graphs/contributors">
  366. <img src="https://contrib.rocks/image?repo=agentscope-ai/ReMe" alt="Contributors" />
  367. </a>
  368. ## 📄 Citation
  369. ```bibtex
  370. @software{ReMe2026,
  371. title = {Remember me, Refine me: Memory Management Kit for Agents},
  372. author = {ReMe Team},
  373. url = {https://reme.agentscope.io},
  374. year = {2026}
  375. }
  376. ```
  377. ## ⚖️ License
  378. This project is open source under the Apache License 2.0. See [LICENSE](./LICENSE) for details.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.