Skip to content
Back to search

mcp-memory

πŸ”₯πŸ–₯️ MCP Memory is a MCP Server that gives MCP Clients (Cursor, Claude, Windsurf and more) the ability to remember information about users (preferences, behaviors) across conversations.

Safety scan

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

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/puliczek--mcp-memory/badge.svg)](/listings/puliczek--mcp-memory)

HTML

<a href="/listings/puliczek--mcp-memory"><img src="/api/listings/puliczek--mcp-memory/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.

1. Install dependencies:

npm install

2. Create a Vectorize index:

npx wrangler vectorize create mcp-memory-vectorize --dimensions 1024 --metric cosine

3. Install Wrangler:

npm run dev

4. Deploy the worker:

npm run deploy

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 Β· README Β· 175 lines

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

  1. <div align="center" >🀝 Show your support - give a ⭐️ if you liked the content
  2. </div>
  3. ---
  4. # MCP Memory
  5. **MCP Memory** is a **MCP Server** that gives **MCP Clients (Cursor, Claude, Windsurf and more)** the **ability to remember** information about users (preferences, behaviors) **across conversations**. It uses vector search technology to find relevant memories based on meaning, not just keywords. It's built with Cloudflare Workers, D1, Vectorize (RAG), Durable Objects, Workers AI and Agents.
  6. ## πŸ“Ί Video
  7. <a href="https://www.youtube.com/watch?feature=player_embedded&v=qfFvYERw2TQ" target="_blank">
  8. <img src="https://github.com/Puliczek/mcp-memory/blob/main/video.png?raw=true" alt="Watch the video" width="800" height="450" border="10" />
  9. </a>
  10. ## πŸš€ Try It Out
  11. ### [https://memory.mcpgenerator.com/](https://memory.mcpgenerator.com/)
  12. ## πŸ› οΈ How to Deploy Your Own MCP Memory
  13. ### Option 1: One-Click Deploy Your Own MCP Memory to Cloudflare
  14. [![Deploy to Cloudflare](https://deploy.workers.cloudflare.com/button)](https://deploy.workers.cloudflare.com/?url=https://github.com/puliczek/mcp-memory)
  15. In **Create Vectorize** section choose:
  16. - **Dimensions:** 1024
  17. - **Metric:** cosine
  18. Click button **"Create and Deploy"**
  19. In Cloudflare dashboard, go to "Workers & Pages" and click on Visit
  20. ![Visit MCP Memory](/visit.png)
  21. ### Option 2: Use this template
  22. 1. Click the "Use this template" button at the top of this repository
  23. 2. Clone your new repository
  24. 3. Follow the setup instructions below
  25. ### Option 3: Create with CloudFlare CLI
  26. ```bash
  27. npm create cloudflare@latest --git https://github.com/puliczek/mcp-memory
  28. ```
  29. ## πŸ”§ Setup (Only Option 2 & 3)
  30. 1. Install dependencies:
  31. ```bash
  32. npm install
  33. ```
  34. 2. Create a Vectorize index:
  35. ```bash
  36. npx wrangler vectorize create mcp-memory-vectorize --dimensions 1024 --metric cosine
  37. ```
  38. 3. Install Wrangler:
  39. ```bash
  40. npm run dev
  41. ```
  42. 4. Deploy the worker:
  43. ```bash
  44. npm run deploy
  45. ```
  46. ## 🧠 How It Works
  47. ![MCP Memory Architecture](/arch.png)
  48. 1. **Storing Memories**:
  49. - Your text is processed by **Cloudflare Workers AI** using the open-source `@cf/baai/bge-m3` model to generate embeddings
  50. - The text and its vector embedding are stored in two places:
  51. - **Cloudflare Vectorize**: Stores the vector embeddings for similarity search
  52. - **Cloudflare D1**: Stores the original text and metadata for persistence
  53. - A **Durable Object** (MyMCP) manages the state and ensures consistency
  54. - The **Agents** framework handles the **MCP protocol** communication
  55. 2. **Retrieving Memories**:
  56. - Your query is converted to a vector using **Workers AI** with the same `@cf/baai/bge-m3` model
  57. - Vectorize performs similarity search to find relevant memories
  58. - Results are ranked by similarity score
  59. - The **D1 database** provides the original text for matched vectors
  60. - The **Durable Object** coordinates the retrieval process
  61. This architecture enables:
  62. - Fast vector similarity search through Vectorize
  63. - Persistent storage with D1
  64. - Stateful operations via Durable Objects
  65. - Standardized AI interactions through Workers AI
  66. - Protocol compliance via the Agents framework
  67. The system finds conceptually related information even when the exact words don't match.
  68. ## πŸ”’ Security
  69. MCP Memory implements several security measures to protect user data:
  70. - Each user's memories are stored in **isolated namespaces** within Vectorize for data separation
  71. - Built-in **rate limiting** prevents abuse (**100 req/min** - you can change it in wrangler.jsonc)
  72. - **Authentication is based on userId only**
  73. - While this is sufficient for basic protection due to rate limiting
  74. - Additional authentication layers (like API keys or OAuth) can be easily added if needed
  75. - All data is stored in Cloudflare's secure infrastructure
  76. - All communications are secured with industry-standard TLS encryption (automatically provided by Cloudflare's SSL/TLS certification)
  77. ## πŸ’° Cost Information - FREE for Most Users
  78. MCP Memory is free to use for normal usage levels:
  79. - Free tier allows 1,000 memories with ~28,000 queries per month
  80. - Uses Cloudflare's free quota for Workers, Vectorize, Worker AI and D1 database
  81. For more details on Cloudflare pricing, see:
  82. - [Vectorize Pricing](https://developers.cloudflare.com/vectorize/platform/pricing/)
  83. - [Workers AI Pricing](https://developers.cloudflare.com/workers-ai/pricing-and-rate-limits/)
  84. - [Workers Pricing](https://developers.cloudflare.com/workers/platform/pricing/)
  85. - [Durable Objects Pricing](https://developers.cloudflare.com/durable-objects/platform/pricing/)
  86. - [Database D1 Pricing](https://developers.cloudflare.com/d1/platform/pricing/)
  87. ## ❓ FAQ
  88. 1. **Can I use memory.mcpgenerator.com to store my memories?**
  89. - Yes, you can use memory.mcpgenerator.com to store and retrieve your memories
  90. - The service is free
  91. - Your memories are securely stored and accessible only to you
  92. - I cannot guarantee that the service will always be available
  93. 2. **Can I host it?**
  94. - Yes, you can host your own instance of MCP Memory **for free on Cloudflare**
  95. - You'll need a Cloudflare account and the following services:
  96. - Workers
  97. - Vectorize
  98. - D1 Database
  99. - Workers AI
  100. 3. **Can I run it locally?**
  101. - Yes, you can run MCP Memory locally for development
  102. - Use `wrangler dev` to run the worker locally
  103. - You'll need to set up local development credentials for Cloudflare services
  104. - Note that some features like vector search or workers AI requires a connection to Cloudflare's services
  105. 4. **Can I use different hosting?**
  106. - No, MCP Memory is specifically designed for Cloudflare's infrastructure
  107. 5. **Why did you build it?**
  108. - I wanted an open-source solution
  109. - Control over my own data was important to me
  110. 6. **Can I use it for more than one person?**
  111. - Yes, MCP Memory can be integrated into your app to serve all your users
  112. - Each user gets their own isolated memory space
  113. 7. **Can I use it to store things other than memories?**
  114. - Yes, MCP Memory can store any type of text-based information
  115. - Some practical examples:
  116. - Knowledge Base: Store technical documentation, procedures, and troubleshooting guides
  117. - User Behaviors: Track how users interact with features and common usage patterns
  118. - Project Notes: decisions and project updates
  119. - The vector search will help find related items regardless of content type
  120. # 🀝 Show your support
  121. <div>🀝 Show your support - give a ⭐️ if you liked the content</div>

Ratings and comments

Sign in to rate this skill.

Sign in

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