# MCP AutoMem [Health: Active]
**Category:** ๐ง Knowledge & Memory
**Repository:** https://github.com/verygoodplugins/mcp-automem
**GitHub Stars:** 64
**Views:** 0
**Installs:** 0
**Upvotes:** 0
**Directory Page:** https://allmcps.com/mcp/mcp-automem
## Description
Graph-vector memory for AI assistants using FalkorDB and Qdrant
## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):
```json
"mcpServers": {
"mcp-automem": {
"command": "npx",
"args": ["-y","@verygoodplugins/mcp-automem"]
}
}
```
## Documentation & README
# AutoMem MCP: Give Your AI Perfect Memory
**One command. Infinite memory. Perfect recall across all your AI tools.**
```bash
npx @verygoodplugins/mcp-automem setup
```
Your AI assistant now remembers everything. Forever. Across every conversation.
https://github.com/user-attachments/assets/fd79112b-5158-4320-a054-8c18ab1ea314
The guided installer โ npx @verygoodplugins/mcp-automem install walks you through local, hosted, or existing-endpoint setup.
Works with **Claude Desktop**, **Cursor IDE**, **Claude Code**, **GitHub Copilot (coding agent)**, **ChatGPT**, **ElevenLabs**, **OpenAI Codex**, **OpenClaw**, **Hermes**, **Grok Build**, **Google Antigravity** - any MCP-compatible AI platform.
## The Problem We Solve
Every AI conversation starts from zero. Claude forgets your coding style. Cursor can't learn your patterns. Your assistant doesn't remember yesterday's decisions.
**Until now.**
AutoMem MCP connects your AI to persistent memory powered by **[AutoMem](https://github.com/verygoodplugins/automem)** - a graph-vector memory service.
## What You Get
### ๐ง Persistent Memory Across Sessions
- AI remembers decisions, patterns, and context **forever**
- Works across **all MCP platforms** - Claude Desktop, Cursor, Claude Code, OpenAI Codex, OpenClaw, Hermes, Grok Build, Google Antigravity
- **Cross-device sync** - same memory on Mac, Windows, Linux
### ๐ Graph-Vector Architecture
- **11 public authorable relationship types** between memories (recall results may also include read-only system/internal relations that are not valid `associate_memories` inputs)
- **Research-validated** approach (HippoRAG 2: 7% better associative memory)
- **Sub-second retrieval** even with millions of memories
### ๐ Works Everywhere You Code
| Platform | Support | Setup Time |
| ------------------ | ------- | ---------- |
| **Claude Desktop** | โ
Full | 30 seconds |
| **Cursor IDE** | โ
Full | 30 seconds |
| **Claude Code** | โ
Full | 30 seconds |
| **GitHub Copilot** | โ
Full | 2 minutes |
| **OpenAI Codex** | โ
Full | 30 seconds |
| **OpenClaw** | โ
Full | 30 seconds |
| **Hermes Agent** | โ
Full | 30 seconds |
| **Grok Build** | โ
Full | 30 seconds |
| **Google Antigravity** | โ
Full | 30 seconds |
| **Any MCP client** | โ
Full | 30 seconds |
## See It In Action
### Claude Desktop with Personal Preferences
_Claude automatically recalls memories using the Personal Preferences template_
### Cursor IDE with Memory Rules
_Cursor uses automem.mdc rule to automatically recall and store memories_
### Claude Code with Session Memory
_Session-start recall plus LLM-judged storage: Claude decides what's durable and stores it via the memory tools_
More platform walkthroughs (Codex, Hermes, Antigravity, remote MCP) live in the **[Installation Guide](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md)**.
## Quick Start
### 1. Set Up AutoMem Service
You need a running AutoMem service (the memory backend). Choose one:
**Option A: Local Development** (fastest, free)
```bash
git clone https://github.com/verygoodplugins/automem.git
cd automem
make dev
```
Service runs at `http://localhost:8001` - perfect for single-machine use.
**Option B: Railway Cloud** (recommended for production)
[](https://github.com/verygoodplugins/automem/blob/main/INSTALLATION.md#railway-deployment)
One-click deploy with $5 free credits. Typical cost: ~$0.50-1/month after trial.
๐ **[AutoMem Service Installation Guide](https://github.com/verygoodplugins/automem/blob/main/INSTALLATION.md)** - Complete setup instructions for local, Railway, Docker, and production deployments.
---
### 2. Install MCP Client
#### Claude Desktop - One-Click Install
Download and double-click to install AutoMem in Claude Desktop:
**[โฌ๏ธ Download AutoMem for Claude Desktop (.mcpb)](https://github.com/verygoodplugins/mcp-automem/releases/latest/download/mcp-automem.mcpb)**
After installing:
1. Claude Desktop will prompt you for your **AutoMem Endpoint** (`http://127.0.0.1:8001` for local)
2. Optionally enter your **API Key** (required for Railway, skip for local)
3. Click Enable
Then add the paste-ready Personal Preferences starter from [`templates/CLAUDE_DESKTOP_INSTRUCTIONS.md`](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/templates/CLAUDE_DESKTOP_INSTRUCTIONS.md). That's it: Claude now has persistent memory and knows when to use it.
#### Other Platforms
Connect your AI tools to the AutoMem service you just started.
```bash
# Guided install - pick where AutoMem runs, verify it, write .env, and
# configure your agents (Codex, Claude Code, Cursor, OpenClaw, Hermes)
npx @verygoodplugins/mcp-automem install
```
Every change is shown in a review plan before anything is written, and each
modified file keeps a `.bak` backup. Add `--dry-run` to preview, `--yes` to
apply non-interactively. See the [Installation Guide](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#guided-install-fastest)
for all flags.
Just need the `.env` + config snippets without the agent setup? Use the lighter wizard:
```bash
# Creates .env and prints config for your AI platform
npx @verygoodplugins/mcp-automem setup
```
**When prompted:**
- **AutoMem Endpoint:** `http://localhost:8001` (or your Railway URL if deployed)
- **API Key:** Leave blank for local development (or paste your token for Railway)
The wizard will:
- โ
Save your endpoint and API key to `.env`
- โ
Generate config snippets for Claude Desktop/Cursor/Code
- โ
Validate connection to your AutoMem service
### 3. Platform-Specific Setup
**For Claude Code (plugin โ recommended):**
```text
# In Claude Code:
/plugin marketplace add verygoodplugins/mcp-automem
/plugin install automem@verygoodplugins-mcp-automem
```
Claude Code prompts for your AutoMem URL and API key at enable time, bundles the MCP server and silent recall/store-tracking hooks, and auto-updates. Prefer hooks and permissions written directly into `~/.claude/` instead? Run `npx @verygoodplugins/mcp-automem claude-code`.
On Windows, the hook payload assumes a POSIX shell environment such as Git Bash, MSYS2, or WSL โ only `bash` is required (the hooks are pure bash+sed).
**For Cursor IDE:**
[](cursor://anysphere.cursor-deeplink/mcp/install?name=memory&config=eyJlbnYiOnsiQVVUT01FTV9BUElfVVJMIjoiaHR0cDovLzEyNy4wLjAuMTo4MDAxIiwiQVVUT01FTV9BUElfS0VZIjoieW91ci1hcGkta2V5LWlmLXJlcXVpcmVkIn0sImNvbW1hbmQiOiJucHggLXkgQHZlcnlnb29kcGx1Z2lucy9tY3AtYXV0b21lbSJ9)
```bash
# Or use CLI to install automem.mdc rule file
npx @verygoodplugins/mcp-automem cursor
```
**Other platforms** โ Claude Desktop (one-click `.mcpb` above, plus the [Personal Preferences template](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/templates/CLAUDE_DESKTOP_INSTRUCTIONS.md)), [OpenAI Codex](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#openai-codex), [Hermes Agent](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#hermes-agent), [Grok Build](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#grok-build), [OpenClaw](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#openclaw), [Google Antigravity](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#google-antigravity), and [GitHub Copilot](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#github-copilot-coding-agent-githubcom):
๐ **[Full Installation Guide](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md)** for every platform's setup and verification steps
---
## Remote MCP via HTTP
An optional sidecar service (deployable to Railway or any Docker host) connects AutoMem to platforms that support remote MCP over **Streamable HTTP** or SSE โ ChatGPT (Developer Mode connectors), Claude.ai web and Claude Mobile, and ElevenLabs Agents.
๐ **[Remote MCP setup](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#remote-mcp-via-http-sidecar)** for deployment, connect URLs, and per-platform screenshots.
## Architecture
```
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ Your AI Platforms โ
โ Claude Desktop โ Cursor โ Claude Code โ
โโโโโโโโโโโโโโโโฌโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ MCP Protocol
โผ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ @verygoodplugins/mcp-automem (this repo) โ
โ โข Translates MCP calls โ AutoMem API โ
โ โข Platform integrations & rules โ
โ โข Handles authentication โ
โโโโโโโโโโโโโโโโฌโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ HTTP API
โผ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ AutoMem Service (separate repo) โ
โ github.com/verygoodplugins/automem โ
โ โโโโโโโโโโโโโโ โโโโโโโโโโโโโโ โ
โ โ FalkorDB โ โ Qdrant โ โ
โ โ (Graph) โ โ (Vectors) โ โ
โ โโโโโโโโโโโโโโ โโโโโโโโโโโโโโ โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
```
**This repo (mcp-automem):**
- MCP client that connects AI platforms to AutoMem
- Platform-specific integrations (Cursor rules, Claude Code hooks, etc.)
- Setup wizards and configuration tools
**[AutoMem service](https://github.com/verygoodplugins/automem):**
- Backend memory service with graph + vector storage
- Deployment guides (local, Railway, Docker, production)
- API server with FalkorDB + Qdrant
## Features
### Core Memory Operations
- **`store_memory`** โ Save memories with content, tags, importance, metadata. Two modes:
- **Single (default)**: top-level `content` plus optional fields, including `embedding`, `t_valid`, `t_invalid`, custom `id`.
- **Batch**: `memories: [...]` (โค500 items) for bulk ingestion. Per-item `id`/`embedding`/`t_valid`/`t_invalid` are not supported in batch mode.
- **`recall_memory`** โ Three modes selected by which params you pass:
- **ID fetch**: `memory_id` โ fetches one memory by ID; updates `last_accessed`.
- **Tag enumeration**: `tags` + `exhaustive: true` โ paginated exact-match listing for cleanup/audit workflows where ranked recall undercounts. Pair with `limit` (โค200) and `offset`; returns `has_more`.
- **Ranked retrieval (default)**: hybrid search across vector, keyword, tags, recency/state controls, score filters, and graph expansion. Supports `state_mode`, `recency_bias`, `scope_fallback`, `expand_respect_tags`, `min_score`, `adaptive_floor`, and diagnostics such as `tag_scope`, `score_filter`, `query_time_ms`, `vector_search`, and per-result `outside_tag_scope`/`state_replaces`.
- **`associate_memories`** โ Create relationships (11 public authorable types; recall results may also include read-only system relations). Supports single-pair mode and batch mode via `associations: [...]` (โค500) with relation-specific props like `reason`, `context`, `resolution`, `observations`, `transformation`, and `role`.
- **`update_memory`** โ Modify existing memories
- **`delete_memory`** โ Two modes:
- **Single (default)**: `memory_id` โ removes one memory and its embedding.
- **Bulk-by-tag**: `tags: [...]` โ bulk-delete all memories matching ANY tag (exact, case-insensitive). No dry-run; verify with `recall_memory({ tags, exhaustive: true })` first.
- **`check_database_health`** โ Monitor service health, degraded state, sync counts, vector dimensions, and enrichment diagnostics when the service provides them
### Advanced Recall (v0.8.0+)
**Multi-hop Reasoning** - Answer complex questions like "What is Amanda's sister's career?"
```javascript
mcp__memory__recall_memory({
query: "What is Amanda's sister's career?",
expand_entities: true, // Finds "Amanda's sister is Rachel" โ memories about Rachel
});
```
**Context-Aware Coding** - Recall prioritizes language and style preferences
```javascript
mcp__memory__recall_memory({
query: "error handling patterns",
language: "typescript",
context_types: ["Style", "Pattern"],
});
```
### Platform Integrations
#### Cursor IDE
- โ
**Memory-first rule file** (`automem.mdc` in `.cursor/rules/`)
- โ
**Automatic memory recall** at conversation start
- โ
**Auto-detects project context** (package.json, git remote)
- โ
**Global user rules option** for all projects
- โ
**Simple setup** via CLI or one-click install
#### Claude Code
- โ
**Native plugin** - MCP server, silent hooks, and skill in one `/plugin install`, with enable-time config prompts and auto-updates
- โ
**LLM-judged storage** - session-start guidance nudges Claude to store, verify, and associate durable memories during normal work
- โ
**Memory rules** in CLAUDE.md guide Claude's memory usage
#### GitHub Copilot
- โ
**Standalone hook JSON files** installed into `$COPILOT_HOME/hooks/` or `~/.copilot/hooks/`
- โ
**Memory rules** template for `copilot-instructions.md`
- โ
**Format flag** - `--format cli` (camelCase) or `--format vscode` (PascalCase)
- โ
**Setup**: `npx @verygoodplugins/mcp-automem copilot --yes`
#### Claude Desktop
- โ
Direct MCP integration
- โ
Paste-ready Personal Preferences starter template
- โ
Full memory API access
## Why AutoMem MCP?
### vs. Building Your Own
- โ
**2 years of R&D** already done
- โ
**Research-validated** architecture (HippoRAG 2, MELODI, A-MEM)
- โ
**Working integrations** across all MCP platforms
- โ
**Active development** and community
### vs. Other Memory Solutions
- โ
**True graph relationships** (not just vector similarity)
- โ
**Universal MCP compatibility** (works with any MCP client)
- โ
**7 memory types** (Decision/Pattern/Preference/Style/Habit/Insight/Context)
- โ
**Self-hostable** ($5/month vs $150+ for alternatives)
### vs. Native AI Memory
- โ
**Persistent across sessions** (not just context window)
- โ
**Cross-platform** (same memory in Claude, Cursor, Code)
- โ
**Structured relationships** (not just RAG)
- โ
**Infinite scale** (no context window limits)
## Documentation
### MCP Client & Integrations (this repo)
- ๐ฆ **[Installation Guide](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md)** - MCP client setup for all platforms
- ๐ **[Remote MCP via HTTP](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#remote-mcp-via-http-sidecar)** - Connect ChatGPT, Claude Web/Mobile, ElevenLabs
- ๐ฏ **[Cursor Setup](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#cursor-ide)** - IDE integration with rules
- ๐ค **[Claude Code Setup](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/templates/CLAUDE_CODE_INTEGRATION.md)** - Plugin install, hooks, and memory rules
- โ ๏ธ **[Deprecations](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/DEPRECATION.md)** - History of the plugin deprecation and its reversal
- ๐ **[OpenAI Codex Setup](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#openai-codex)** - Codex CLI/IDE/Cloud integration
- ๐ช **[Google Antigravity Setup](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#google-antigravity)** - Raw MCP config via Antigravity's MCP Store
- ๐ **[MCP Tools Reference](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/INSTALLATION.md#mcp-tools)** - All memory operations
- ๐ **[Changelog](https://github.com/verygoodplugins/mcp-automem/blob/HEAD/CHANGELOG.md)** - Release history
### AutoMem Service (separate repo)
- ๐๏ธ **[AutoMem Service](https://github.com/verygoodplugins/automem)** - Backend repository
- ๐ **[Service Installation](https://github.com/verygoodplugins/automem/blob/main/INSTALLATION.md)** - Local, Railway, Docker deployment
- โ๏ธ **[API Documentation](https://github.com/verygoodplugins/automem#api-reference)** - REST API reference
- ๐งช **[Evaluation Lab](https://github.com/verygoodplugins/automem-evals)** - Exploratory recall-quality benchmarks and ruleset A/B testing
## The Science Behind AutoMem
The AutoMem service implements cutting-edge 2025 research:
- **[HippoRAG 2](https://arxiv.org/abs/2502.14802)** (OSU, June 2025): Graph-vector approach achieves 7% better associative memory
- **A-MEM** (July 2025): Dynamic memory organization with Zettelkasten principles
- **MELODI** (DeepMind, 2025): 8x memory compression without quality loss
- **ReadAgent** (DeepMind, 2024): 20x context extension through gist memories
This MCP package provides the bridge between your AI and that research-validated memory system. The backend has also been benchmarked on the neutral [Agent Memory Benchmark](https://automem.ai/benchmarks), including BEAM large-context scaling tiers โ reproducible end to end, so you can run it yourself.
## Community & Support
- ๐ฌ **[Discord](https://automem.ai/discord)** - Join the community, get help, share feedback
- ๐ฆ **[X Community](https://x.com/i/communities/2013114118912225326)** - Discussion and updates
- ๐ฃ **[@automem_ai](https://x.com/automem_ai)** - Official announcements
- ๐ฆ **[NPM Package](https://www.npmjs.com/package/@verygoodplugins/mcp-automem)** - This MCP client
- ๐ฌ **[AutoMem Service](https://github.com/verygoodplugins/automem)** - Backend repo with deployment guides
- ๐ **[GitHub Issues](https://github.com/verygoodplugins/mcp-automem/issues)** - Bug reports and feature requests
## Contributing
We welcome contributions! Please:
1. Fork the repository
2. Create a feature branch
3. Make your changes with tests
4. Submit a pull request with a Conventional Commit title such as `fix:`, `feat:`, `docs:`, or `chore:`
5. Do not prefix the PR title with labels like `[codex]` or `[wip]` because the squash-merge commit is taken from the PR title
## License
MIT - Because great memory should be free.
---
**Ready to give your AI perfect memory?**
```bash
npx @verygoodplugins/mcp-automem setup
```
_Built with obsession. Validated by neuroscience. Powered by graph theory. Works with every MCP-enabled AI._
_Designed by Jack Arturo at [Very Good Plugins](https://verygoodplugins.com)_ ๐งก
**Transform your AI from a tool into a teammate. Start now.**