Research & Knowledge Management

Automated Research & Knowledge Agent

Memory + Web Search/Fetch + Obsidian Note Storage

A continuous research assistant that searches the web, synthesizes technical topics, stores persistent notes, and remembers user preferences across sessions.

Included MCP Servers (2)

  • Server Memory MCP

    Persistent knowledge graph and entity memory across chat sessions.

    npx -y @modelcontextprotocol/server-memory
  • Search & Fetch MCP

    Scrapes web pages, docs, and search results.

    npx -y @modelcontextprotocol/server-fetch

Optimized Agent System Prompt

Paste this system prompt into Cursor (.cursorrules), Claude Desktop, Windsurf, or Antigravity:

Code
You are an elite Knowledge Researcher and Technical Analyst. You possess long-term memory graph access and web data extraction capabilities over Model Context Protocol (MCP).

Your research procedure:
1. Query your Memory MCP graph at the start of every request to check for user context, past findings, or established preferences.
2. When researching new topics, fetch authoritative sources using Fetch MCP.
3. Extract core technical facts and store synthesized knowledge entities back into your Memory MCP graph so knowledge accumulates permanently across sessions.

Combined Suite `claude_desktop_config.json`

One single JSON configuration containing all required MCP servers for this workflow:

JSON Config
{
  "mcpServers": {
    "modelcontextprotocol-server-memory": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-memory"
      ]
    },
    "modelcontextprotocol-server-fetch": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-fetch"
      ]
    }
  }
}

Frequently Asked Questions

Server Memory MCP stores JSON entities locally on your machine, completely private and offline.