MemoryPlugin

MemoryPlugin is an MCP server that provides AI assistants with long-term memory, cross-session context, and knowledge retrieval across multiple platforms. It connects MCP-compatible clients such as Claude Desktop, VS Code, GitHub Copilot Chat, Cursor, and Windsurf to a hosted cloud service without requiring local background runtimes. Software developers, researchers, and knowledge workers use the integration to maintain conversational continuity, store persistent preferences, and index important project documentation. The server enables models to dynamically store and recall facts across customized buckets, summarize previous discussions, search through imported chat histories, and retrieve excerpts from uploaded files. Operating over standard HTTPS or Server-Sent Events with OAuth 2.0 PKCE authentication and Dynamic Client Registration, it eliminates the need to manage manual API keys in client configuration files. By supplying automated summaries and cross-conversation context, it prevents users from needing to re-explain project constraints, personal workflows, and established facts across separate AI chat sessions.

Category: AI Memory & Context

Tags: context, memory, personalization, recall

Visit MemoryPlugin

How to install and configure MemoryPlugin

1. Open your MCP client configuration file, such as claude_desktop_config.json for Claude Desktop or .vscode/mcp.json for VS Code. 2. Add the memoryplugin-remote configuration entry under mcpServers: json { "mcpServers": { "memoryplugin-remote": { "type": "http", "url": "https://www.memoryplugin.com/api/mcp/mcp" } } } 3. Restart or reload your client. 4. Complete the browser-based OAuth 2.0 sign-in flow when prompted by the client to approve access and connect your MemoryPlugin account.

What you can do with MemoryPlugin

• Storing coding conventions, architecture decisions, and project constraints so coding assistants remember guidelines across development sessions. - Retrieving relevant facts and previous discussion points across multiple chats without manually copying and pasting conversation history. - Searching through imported chat histories and uploaded reference documents to extract context directly inside client interfaces. - Maintaining user preferences, operational guidelines, and domain-specific knowledge across different AI clients using a unified memory store.

Key facts

• AI Memory & Context, Notes & Knowledge Management
• context, memory, personalization, recall

Part of MCP Servers

What can MemoryPlugin do?

MemoryPlugin provides AI clients with persistent memory, bucket-based information storage, smart categorization, and chat history search. It also enables file search across uploaded documents and provides an automatic chat history overview to supply LLMs with rich background context.

Which MCP clients work with MemoryPlugin?

MemoryPlugin works natively with HTTP-capable and SSE-capable clients including Claude Desktop, VS Code with GitHub Copilot Chat, Cursor, and Windsurf. For clients lacking native remote HTTP support, it can connect via a local proxy utility like mcp-remote.

How does authentication work on the Remote MCP server?

The remote server uses OAuth 2.0 with PKCE and supports Dynamic Client Registration. When connecting for the first time, your client opens a browser consent screen to sign in. Tokens are handled automatically without needing to paste long-lived API keys into configuration files.

Does my client email need to match my MemoryPlugin email?

No. Your MemoryPlugin account email operates independently and does not need to match the email address associated with your Claude, ChatGPT, or IDE accounts. Permissions and storage access are linked directly to your MemoryPlugin account during the OAuth approval process.