Memex Targeted Search Server

Provides targeted search across Memex conversations and project files via MCP, returning concise snippets and relevant results.

typescript

0

GitHub Stars

typescript

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "memextech-memex-targeted-search-server": {
      "command": "node",
      "args": [
        "/path/to/memex-targeted-search-server/dist/index.js"
      ]
    }
  }
}

You can query Memex conversations and project files through a targeted search MCP server. It returns precise snippets and relevant results to help agents quickly locate commands, conversations, and code across your history and workspace.

How to use

Use an MCP client to connect to the Memex Targeted Search MCP Server and invoke its search functions. You can search conversation history, fetch targeted conversation snippets, search project files by content and type, get summaries of a project with detected technologies, and locate specific commands or CLI usage from past conversations. The server prioritizes relevance, offers faceted filtering (dates, projects, file types), and returns concise snippets to keep context focused while allowing you to drill down for details.

How to install

Prerequisites you need before starting: node and npm are installed on your system.

Clone the project, install dependencies, build, and start the server locally.

Configuration and usage notes

Configure the MCP client to connect to the Memex Targeted Search MCP Server using a local stdio endpoint. The following configuration runs the server locally via Node.js and points to the built entry file.

Troubleshooting

If the server cannot start, verify that Node.js and npm are installed, the build completed without errors, and the path to dist/index.js is correct. Ensure the required history and workspace directories exist and are accessible.

Available tools

search_conversations

Search conversation history by text, metadata, and filters to surface relevant conversations.

get_conversation_snippet

Retrieve specific parts of conversations to provide concise context without overload.

search_projects

Search project files by content, file types, and names to locate code and docs.

get_project_overview

Analyze project structure and detect technologies to summarize a workspace.

find_command

Find exact commands, CLI usage, or code snippets from conversation history.