Memex Targeted Search Server

Memex Targeted Search Server is an MCP server that provides targeted retrieval across local Memex conversation records and project workspaces. It connects directly to local application storage and development directories, allowing AI clients such as Claude Desktop to locate past conversations, code snippets, shell commands, and project structures. Software developers, researchers, and power users who accumulate extensive chat histories and multi-project repositories use this server to retrieve contextual information without exhausting context window limits. Instead of loading entire chat logs into the model context, it extracts concise snippets, executes scoped file searches, filters conversations by metadata or date ranges, and inspects workspace directory layouts. This targeted approach enables assistants to recover forgotten CLI syntax, inspect implementation patterns across projects, and recall historical development decisions while keeping prompt tokens focused and efficient.

Category: AI Memory & Context

Tags: context, history, memex, retrieval, search

Visit Memex Targeted Search Server

How to install and configure Memex Targeted Search Server

1. Clone the repository: bash git clone https://github.com/memextech/memex-targeted-search-server.git cd memex-targeted-search-server 2. Install dependencies: bash npm install 3. Build the server: bash npm run build 4. Add the server to your MCP client configuration (such as Claude Desktop config): json { "mcpServers": { "memex-search": { "command": "node", "args": ["/path/to/memex-targeted-search-server/dist/index.js"] } } } 5. Restart your MCP client to initialize the server tools.

What you can do with Memex Targeted Search Server

• Retrieve forgotten CLI commands, deployment scripts, or shell syntax stored across previous Memex chat sessions. * Search historical conversations by date, project name, or keywords to recover past implementation discussions. * Pull specific conversation message ranges without importing large chat files directly into the LLM context. * Query project directories for specific code keywords, imports, or file extensions across local workspace folders. * Generate structural summaries and technology stack overviews for specific project repositories in the workspace.

Key facts

• https://github.com/memextech/memex-targeted-search-server
• AI Memory & Context, Notes & Knowledge Management, Web Search & Research
• context, history, memex, retrieval, search

Part of MCP Servers

What is Memex Targeted Search Server?

Memex Targeted Search Server is an Model Context Protocol server built with TypeScript and Node.js. It gives AI assistants tools to search through local Memex chat history and workspace project directories to retrieve snippets, commands, and file references without overloading the context window.

What can Memex Targeted Search Server do?

The server exposes tools to search conversation records, retrieve indexed message snippets, locate specific CLI or code commands, inspect file content filtered by file extension, and analyze project folder structures and technology stacks.

Where does the server look for files by default?

The server searches conversation history in the local Memex application directory at ~/Library/Application Support/Memex/history/ and project files located in the ~/Workspace/ directory.

Which MCP clients work with Memex Targeted Search Server?

It works with any Model Context Protocol compliant client, including Claude Desktop, by launching the compiled Node.js script through standard local process transport.

What license does Memex Targeted Search Server use?

According to the repository documentation, Memex Targeted Search Server is open-source software released under the ISC License.