ReMe

A local-first, self-evolving personal knowledge base for AI agents.

Previous versions: 0.3.x · 0.2.x · MemoryScope

✨ Why ReMe?

🧠 ReMe turns conversations and resources into readable, editable, searchable, and interconnected Markdown memory. Agents such as QwenPaw and DeepSeek Harness can share the same workspace to retrieve, maintain, and evolve knowledge, while users retain control of the durable files.

• Memory as File, File as Memory: ReMe stores durable memory as ordinary Markdown with frontmatter and wikilinks. Users and agents can inspect, edit, move, sync, and back it up with familiar tools, while indexes and generated metadata remain rebuildable.
• Self-evolving knowledge base: ReMe progressively turns conversations and resources into daily notes and long-term knowledge, preserving sources while refining facts, preferences, procedures, and relationships over time.
• Recall is precise and context-aware. BM25, optional embeddings, and wikilink expansion retrieve relevant line-level passages and their relationships without loading the entire knowledge base into the agent context.
• One memory workspace works across agents. Personal assistants, coding agents, and other agent runtimes can share the same local workspace through native integrations, SKILL.md, CLI, HTTP, MCP, or Python APIs.

📰 Latest Updates

• [2026.10] - Decision-model context compaction: score old tool calls and results against the next action, preserve protected history, and return a recoverable compaction proposal. Includes an SVG walkthrough, two live-model integration scenarios, and a 14-query fixed-summary Top-N benchmark with a forced Hermes summary baseline.

• [2026.10] - ReMe Multimodal Memory update: Auto Memory reads images with their conversation context and preserves image sources; Auto Resource turns image files into searchable daily cards linked to the originals.

• [2026.10] - ReMe Studio Playground is live: explore example memory files, edit Markdown, and browse linked memory graphs right in your browser—no installation or backend required. Everyone is welcome to try it out!

  

• [2026.09] - ReMe Memory Tags published: an introduction to file-native entity tags, rebuildable tag indexes, and tag-filtered memory search.

• [2026.09] - Hermes Agent memory provider available: choose HTTP or embedded mode for automatic recall before model calls and asynchronous auto_memory after completed turns. The integration supports Hermes Agent 0.21+ and includes profile-aware background work.

More updates

• [2026.09] - OpenClaw plugin released: install it from ClawHub or npm to add native memory recall, automatic conversation capture, and scheduled consolidation to OpenClaw.

• [2026.09] - DeepSeek Harness plugin released: install it from Awesome DSH Plugin or npm for long-term-memory guidance, reme_search, automatic memory, Auto Dream, and ReMe Status.

• [2026.08] - ReMe blog published: the ReMe blog introduces the local-first memory architecture, self-evolving workflows, hybrid search, proactive discovery, and benchmark results.

• [2026.08] - New ReMe ecosystem plugins: Daily Paper discovers and analyzes papers and generates file-native briefs, while Auto Fin researches the latest 24 hours of topic-related CLS news and builds traceable reports with local memory. Try them out.

• [2026.08] - Plugin development support released: use Plugin Development and Plugin Management to extend ReMe with Components, Steps, and Jobs. Contributions and new community plugins are welcome.

• [2026.08] - ReMe’s experience-driven enhancement method for agent tool use is available on arXiv:2608.03403.

• [2026.07] - Our paper Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution has been accepted to Findings of ACL 2026.

🚀 Quick Start

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.

Optional decision models use pip install "reme-ai[core,decision]". Configure a TypeSafe API key and run the service demo as described in Decision model configuration.

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 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, then change components.file_store.default.embedding_store from "" to default. See the memory search guide for details.

Start the Service

Start ReMe in one terminal and leave it running:

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

From another terminal, call the running service. The CLI discovers its configuration; the curl example uses the default port 2333:

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 for source builds, configuration, and development.

🤝 Use ReMe with Your Agent

ReMe can run as a local memory service accessed through the CLI, HTTP API, or MCP server, or it can be embedded in the host process through its Python API. Host integrations can add memory guidance, recall, and capture to the agent lifecycle according to the capabilities of each runtime.

Agent	Recommended path	Available after integration
DeepSeek Harness	Install @agentscope-ai/reme-dsh-plugin 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.
OpenClaw	Install @agentscope-ai/reme-openclaw-plugin with openclaw plugins install clawhub:@agentscope-ai/reme-openclaw-plugin.	Native memory tools, recall before user-triggered runs, and automatic turn capture.
QwenPaw	Embed ReMe in-process through its Python API.	Reuse the host lifecycle and model config while keeping memory local and file-based.
Claude Code	Start ReMe, then install the Claude Code plugin.	Automatic recall, batched completed-turn recording, and MCP tools.
Qoder	Start ReMe, then install the Qoder plugin.	Automatic recall, retryable completed-turn recording, MCP tools, and the bundled skill.
Hermes	Install the ReMe provider and choose HTTP or embedded mode.	Recall before model calls and asynchronous auto_memory after each completed turn.
Codex	Start ReMe, then install the Codex plugin.	Native hooks for automatic recall and batched recording, plus MCP tools.
Other CLI agents	Install or copy the ReMe Memory skill.	Search, read, and write memory through the CLI.

Integration demos

	Auto Memory	Auto Dream
QwenPaw
Claude Code

🧠 How ReMe Works

Memory as File, File as Memory.

ReMe treats memory as files, progressively processing filtered conversation source records and external resources from session/ and resource/ into daily/, then digest/. The default workspace is .reme/ under the current directory; workspace_dir=... selects a different user-owned location.

Workspace Layout

<workspace_dir>/
├── metadata/       # Rebuildable indexes, graphs, catalogs, and caches
├── session/        # Conversation source records and agent sessions
│   ├── dialog/
│   │   └── <session_id>.jsonl  # Source messages saved by auto_memory
│   └── claude_code/
│       └── <session_id>.jsonl  # ReMe copy used by auto_memory_cc
├── mem_session/    # Generated agent-wrapper sessions/config, not user memory
│   ├── agentscope/
│   ├── claude_config/
│   └── codex/
├── resource/            # External raw materials
│   ├── <resource>.<ext>  # Root-level files enter today's daily layer
│   └── YYYY-MM-DD/
│       └── <resource>.<ext>
├── daily/               # Lightly processed memory: daily facts, conversation summaries, resource readings
│   ├── YYYY-MM-DD.md
│   └── YYYY-MM-DD/
│       ├── <generated_name>.md  # Topic-named conversation or resource card
│       └── interests.yaml
└── digest/              # Long-term memory: personal facts, procedural experience, knowledge nodes
    ├── personal/
    │   └── {topic/event}.md
    ├── procedure/
    │   └── {topic/event}.md
    └── wiki/
        └── {topic/event}.md

Memory Lifecycle

ReMe follows a capture → index → consolidate → recall loop. Workspace files remain the durable source of truth; everything under metadata/ is rebuildable.

Capability	Entry point	What it does	Output
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
auto_resource	Resource watcher or reme auto_resource	Turns files under resource/ into source-linked, content-named daily cards.	daily/<date>/<resource-card>.md
auto_index	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
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/**
proactive_read	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

Search returns matching chunks with line ranges and bounded wikilink neighbors. Optional vector results are fused with BM25 through reciprocal rank fusion (RRF).

📊 Benchmarks

ReMe evaluates multi-session and long-context memory with agentic search-and-read workflows. The figures below are the published reference runs in this repository; model, prompt, dataset, and judging details are documented with each benchmark.

Benchmark	Setting	Sample size	Agentic score	Focus
LongMemEval cleaned-s	Overall	500 questions	89.4%	Cross-session retrieval, knowledge updates, and temporal reasoning
BEAM	100K context	20 cases / 400 questions	66.1%	Ten types of long-context memory tasks
BEAM	1M context	35 cases / 700 questions	65.0%	Ultra-long conversation settings

ReMe also achieved a 0.580 PROC score across five user personas in the repository’s π-Bench evaluation, 2.4% above NanoBot under the same test-model configuration. PROC measures proactive handling of hidden intent, clarification, cross-session preferences and conventions, task dependencies, and underspecified requests.

🧩 Extensions and Plugins

Plugins are optional Python distributions that contribute Component, Step, or Job backends and configuration. They are installed separately and enabled explicitly by configuration. Daily Paper and Auto Fin are independently packaged plugins; see their documentation for Daily Paper and Auto Fin.

Plugin	Capability
Daily Paper	Discover and rank papers, analyze PDFs with an agent, and generate file-native notes and a five-minute brief.
Auto Fin	Fetch topic-related CLS news, search ReMe history, and generate wikilink-backed Markdown reports.

See Plugin Management to install, inspect, validate, enable, and uninstall ReMe plugins.

📚 Documentation

These guides cover the main user workflows and the runtime contracts implemented by the current code.

Guide	What you will learn
Quick Start	Install ReMe, start the service, and run the first file and memory operations.
Configuration	Configure the workspace, models, Service, Jobs, Components, plugins, and CLI overrides.
Services and Deployment	Use HTTP, SSE, MCP, and Studio while respecting the default security boundary.
Memory as File	Understand workspace layers, frontmatter, wikilinks, chunks, and the file-as-source-of-truth model.
Auto Memory	Preserve source conversations and distill reusable daily memory cards.
Auto Resource	Import supported text and image resources as source-linked daily cards.
Auto Dream and Auto Link	Consolidate daily notes into evolving digest nodes and readable wikilink relationships.
Memory Search	Use BM25, optional vectors, RRF fusion, line-range recall, and progressive link expansion.
Proactive	Read interest topics safely and integrate them into a host agent’s decision flow.
Application Scenarios	Follow concrete financial research, coding-memory, and personal knowledge-base examples.
Framework	Understand Application, Job, Step, Component, service, configuration, and lifecycle boundaries.
Agent Integrations	Choose an interface and connect DSH, Claude Code, Codex, Qoder, OpenClaw, Hermes, or another agent.
DSH plugin, Claude Code plugin, and Qoder plugin	Configure host-native recall, automatic capture, consolidation, and diagnostics.
CLI and Job API	Learn command syntax and use the generated default Job parameter reference.
Operations and Recovery	Diagnose services, maintain indexes, and back up, migrate, or recover a workspace.
ReMe Blog	Read the product story, design rationale, examples, and benchmark summary.

🛠️ Common Commands

Run reme help for the full job list. Common workspace and maintenance commands are:

Command	Purpose
reme status	Show stateful data-component memory estimates and process RSS.
reme search	Retrieve memory with BM25 and wikilinks by default, plus vectors when enabled.
reme read / reme write / reme edit	Inspect and maintain Markdown memory files.
reme traverse / reme graph_snapshot	Explore wikilink neighborhoods or the category-rooted digest graph.
reme chat	Stream a read-only, workspace-aware agent conversation. Requires LLM credentials.
reme reindex	Rebuild BM25 and embedding indexes from already-ingested chunks.

🤝 Community and Contributing

• Issues, requests, and help: Check Open Issues first. If there is no related discussion, open one with the background, expected behavior, and impact scope.
• Code contributions: Before making changes, read the repository’s contribution guide. Source, schemas, and tests are the authoritative architecture and extension guide.
• Documentation contributions: Update the canonical files under docs/en/, docs/zh/, or the relevant package directory in this repository. The documentation site is generated from these files.
• Commit convention: Conventional Commits are recommended, for example feat(search): add link expansion option or docs(zh): update quick start.
• Pre-submit checks: Before submitting a PR, try to run pre-commit run --all-files and pytest. If tests that depend on LLMs, embeddings, or external services cannot run, explain that in the PR.
• Documentation: Visit reme.agentscope.io.

Contributors

Thanks to everyone who has contributed to ReMe:

📄 Citation

@software{ReMe2026,
  title = {Remember me, Refine me: Memory Management Kit for Agents},
  author = {ReMe Team},
  url = {https://reme.agentscope.io},
  year = {2026}
}

⚖️ License

This project is open source under the Apache License 2.0. See LICENSE for details.