Quick Start

This path gets one thing working first: install the LTM server, register it with your MCP client, index a folder, then search it from your agent. STM is optional and can be added after the basic memory loop works.

1. Install

uv tool install 'memtomem[all]'       # or: pipx install 'memtomem[all]'
mm --version

[all] includes the Web UI, local ONNX embeddings, code chunking, Korean tokenizer support, and provider clients for Ollama / OpenAI. For a smaller BM25-only install, see Installation.

If mm --version prints an older version right after install, upgrade with refreshed package metadata:

uv tool install 'memtomem[all]' --refresh

2. Run Setup

mm init

The wizard starts with a preset picker:

Preset	Best for	What it does
Minimal	First smoke test, smallest install	BM25 keyword search only
English	Most English projects	Local ONNX embeddings + English reranker
Korean	Korean or multilingual notes	Multilingual embeddings, reranker, Korean tokenizer

For scripts or CI:

mm init -y                   # minimal preset, no prompts
mm init --preset english -y  # English preset, no prompts
mm init --preset korean -y   # Korean preset, no prompts
mm init --advanced           # full wizard

During setup you can also enroll existing AI-agent memory folders, such as Claude Code memories, Claude plans, or Codex CLI memories. Enrollment only registers the folders for future watching; seed existing files once with mm index.

3. Check Locally

Before wiring an editor, confirm the CLI can open the config and database:

mm status

You should see storage, embedding, and chunk-count information. Zero chunks is normal before indexing.

4. Connect Your MCP Client

If you picked “auto-register with Claude Code” in the mm init wizard, LTM (memtomem-server) is already registered — no extra LTM command needed.

For a manual registration (skipped the wizard prompt, or moved to a new machine):

claude mcp add memtomem -s user -- memtomem-server

Ask your agent to “call mem_status” after registration. That verifies the MCP client is launching the same server your terminal sees.

5. Add and Search Memory

Use the CLI once so you can see the full loop without depending on an agent:

mm add \
  "The staging deploy uses the blue-green checklist in docs/deploy.md" \
  --tags "deploy,ops"
mm search "staging deployment checklist"

Then index the notes, docs, or code you want agents to search:

mm index ~/notes
mm index ~/projects/my-app/docs

After that, use natural language in your MCP client:

Tell your agent	Tool usually called	Result
”Check memtomem status”	`mem_status`	Connection and index summary
”Remember this decision…”	`mem_add`	New memory stored in markdown
”Index this docs folder”	`mem_index`	Existing files become searchable
”Search for the deploy checklist”	`mem_search`	Ranked snippets from your memory

6. Open the Web UI

mm web --open

The default Web UI is the polished user surface for search, sources, tags, timeline, settings, and Context Gateway. mm web --dev adds maintainer-oriented pages that most users do not need.

Optional: Add STM Later

memtomem-stm is a separate proxy for proactive memory surfacing and response compression. Add it after the LTM flow above works:

uv tool install memtomem-stm
mms init --mcp claude
mms health

See STM Overview when you want tool-response compression or automatic memory injection.

Memory Persistence — save in one session, recall in another
Installation — install options and extras
Hybrid Search — BM25 + vector + RRF fusion
MCP Tools — core, standard, and full tool modes
Context Gateway — sync agents, skills, and commands across runtimes