Overview

The Cameron Rohn enables AI models to perform advanced topic research using web search and LLM synthesis. It's part of the Model Context Protocol (MCP) system, providing a safe and standard way to connect AI with research capabilities. For more advanced memory capabilities, consider integrating with a Knowledge Graph Memory or Supavec. You can also explore other AI/ML tools for enhanced functionality.

Created by:

Developed by

Key Features

🔍

Web Search Integration

Utilize web search APIs (Tavily, Perplexity) to gather up-to-date information.

🧠

LLM Synthesis

Synthesize research findings using local Ollama LLMs for comprehensive answers.

🔄

Iterative Research Process

Iteratively improve summaries by identifying knowledge gaps and generating new search queries.

📝

Markdown Summary with Sources

Provides a final markdown summary of research with all sources used.

Available Tools

Quick Reference

Tool	Purpose	Category
`research`	Research a given topic	Research
`get_status`	Get the status of ongoing research	Utility
`configure`	Configure research parameters	Configuration

Detailed Usage

research▶

Initiate a deep research process on a specified topic.

use_mcp_tool({
  server_name: "ollama-deep-researcher",
  tool_name: "research",
  arguments: {
    topic: "The impact of AI on climate change"
  }
});

Returns a research ID to track progress.

get_status▶

Retrieve the current status and results of an ongoing research task.

use_mcp_tool({
  server_name: "ollama-deep-researcher",
  tool_name: "get_status",
  arguments: {
    researchId: "research-12345"
  }
});

Returns the current status, progress, and final summary if completed.

configure▶

Configure parameters for the research process, such as max loops, LLM model, and search API.

use_mcp_tool({
  server_name: "ollama-deep-researcher",
  tool_name: "configure",
  arguments: {
    maxLoops: 5,
    llmModel: "deepseek-r1:8b",
    searchApi: "tavily"
  }
});

Updates the research configuration for subsequent tasks.

Installation

{
  "mcpServers": {
    "ollama-deep-researcher": {
      "command": "node",
      "args": [
        "path/to/mcp-server-ollama-deep-researcher/build/index.js"
      ],
      "env": {
        "TAVILY_API_KEY": "your_tavily_key",
        "PERPLEXITY_API_KEY": "your_perplexity_key",
        "LANGSMITH_API_KEY": "your-langsmith-key"
      }
    }
  }
}

Environment Variables:

Ensure TAVILY_API_KEY, PERPLEXITY_API_KEY, and LANGSMITH_API_KEY are set in your environment or directly in the env object.

Common Use Cases

1. Automated Literature Review

Perform systematic literature reviews on specific topics, aggregating information from various web sources.

2. Knowledge Gap Analysis

Identify missing information or inconsistencies in existing knowledge bases by iteratively researching and synthesizing findings.

3. Research Synthesis for Reports

Generate comprehensive summaries and reports on complex subjects, complete with citations and source validation.

Sources

Fetch MCP Server

Fetch MCP servers enable AI models to retrieve and process content from web pages, converting HTML to markdown for easier consumption.

Filesystem MCP Server

Filesystem MCP servers enable AI models to interact with local file systems, providing capabilities for file operations, directory management, and secure file access within specified boundaries.

MongoDB MCP Server

MongoDB MCP servers enable AI models to interact with MongoDB databases and MongoDB Atlas, providing capabilities for document operations, aggregation pipelines, cloud database management, and natural language queries.

Ollama Deep Researcher: AI Model for Web Search & LLM Synthesis