Quick Start

This page uses ReMe’s model-free capabilities to complete a verifiable write → index → search → read loop.

1. Install

ReMe requires Python 3.11 or newer:

pip install "reme-ai[core]"

For source development:

git clone https://github.com/agentscope-ai/ReMe.git
cd ReMe
pip install -e ".[core]"

2. Start the Service

In one terminal, run:

reme start

The default workspace is .reme/ under the launch directory, and the service listens on 127.0.0.1:2333. In another terminal, check it:

reme version
reme health_check
reme help

Use another port when necessary:

reme start service.port=8181

When the package includes the web build, open http://127.0.0.1:2333/ for ReMe Studio. It can browse, edit, and search the workspace, chat with the read-only workspace agent, and inspect the digest wikilink graph. Set service.web_enabled=false to disable it, or use service.web_static_dir / REME_WEB_STATIC_DIR for a custom build. The Job API still works when no web build is found.

3. Write a Memory

Use reme write to create a Markdown memory in the durable digest/ area:

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

ReMe stores agent memory as user-readable Markdown."

ReMe adds the .md suffix and writes frontmatter plus content to .reme/digest/wiki/quick-start-demo.md.

4. Search and Read

The background index watches Markdown under daily/ and digest/, so it normally discovers the file automatically. If it does not appear immediately, wait briefly or run reme reindex once. The full rebuild also scans resource/ and JSONL.

reme search query="agent memory Markdown" limit=5
reme read path=digest/wiki/quick-start-demo.md

Search also accepts an inclusive date range:

reme search query="agent memory Markdown" start_date=2026-08-01 end_date=2026-08-12 limit=5

Open the file in the workspace too. Matching CLI output and disk content confirm that the minimum loop works. After completing the loop, the workspace contains these top-level areas:

.reme/
├── metadata/    derived indexes, graph, and runtime state
├── session/     source conversations
├── mem_session/ generated agent-wrapper sessions and configuration
├── resource/    source resources
├── daily/       daily refinement
└── digest/      durable memory

Conversation refinement, resource interpretation, and consolidation require an LLM. Semantic search also requires embeddings. Add a .env in the launch directory when you need them:

LLM_API_KEY=sk-xxx
LLM_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1

EMBEDDING_API_KEY=sk-xxx
EMBEDDING_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1

Model names and embedding-store activation are configuration choices. The authoritative defaults are in reme/config/default.yaml. Never commit secrets from .env.

Compare the available integration paths, then use the Python SDK or follow an agent integration guide.
Find workspace, configuration, and interface details in the ReMe reference.
Review the ReMe FAQ for connection or indexing problems.

Get Started

Integration

Resources

1. Install

2. Start the Service

3. Write a Memory

4. Search and Read

5. Enable Memory Refinement When Needed

Next

​1. Install

​2. Start the Service

​3. Write a Memory

​4. Search and Read

​5. Enable Memory Refinement When Needed

​Next

1. Install

2. Start the Service

3. Write a Memory

4. Search and Read

5. Enable Memory Refinement When Needed

Next