Chronulus AI Forecasting

MCP server integrating Chronulus AI prediction agents with Claude

Artificial intelligence chatbots Research and data #AI prediction #Claude integration #MCP protocol #Data analysis .Python

rating : 2.5 points

downloads : 7.4K

update time : 2025-04-28

Open Site

What is the Chronulus MCP Server?

The Chronulus MCP Server allows you to interact with Chronulus AI's prediction and time series analysis agents through the Claude desktop client. It supports the integration of multiple tools and services, providing you with powerful data analysis capabilities.

How to use the Chronulus MCP Server?

First, install the Claude desktop client, and then configure the Chronulus MCP Server. You can choose different installation methods (such as pip, Docker, or uvx) and add it to your Claude configuration file.

Applicable Scenarios

Suitable for enterprise and individual users who need to conduct time series analysis, predictive modeling, and data visualization. Particularly suitable for fields such as finance and supply chain management.

Main Features

Multi - tool Integration

Supports integration with third - party tools such as Fetch and Filesystem, facilitating users to handle different types of data sources.

Advanced Prediction Algorithms

Built - in multiple advanced machine learning models for generating accurate time series predictions.

Custom Chart Generation

Dynamically generate high - quality charts according to user needs, accompanied by detailed explanations.

Advantages

Easy to integrate into existing workflows

Provide highly customizable prediction solutions

Limitations

High security requirements for API keys

Some complex tasks may require additional computing resources

How to Use

Download and install the Claude desktop client

Visit the Claude official website to download the version suitable for your operating system.

Get the Chronulus API key

After logging in to your Chronulus account, find the API key on the personal settings page.

Configure the Chronulus MCP Server

Add the Chronulus server to the Claude configuration file.

Usage Examples

Time Series Prediction

Use the Chronulus MCP Server to predict future trends based on historical sales data.

File Import and Analysis

Upload a PDF document to Claude, and use the Chronulus agent to extract key information and generate a summary.

Frequently Asked Questions

How to ensure the security of my API key?

Why do some commands not work properly?

Related Resources

Chronulus Official Documentation

Gain in - depth knowledge of Chronulus's features and services.

Claude Community Forum

Exchange experiences and technologies with other users.

🚀 Quick Start: Claude Desktop Version

📦 Installation

Claude Desktop is currently available for macOS and Windows systems.

Download and install Claude Desktop from here.

⚙️ Configuration

Configure the Claude desktop client following the general instructions provided here.

You can find your Claude configuration file at the following locations:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Then, add the appropriate configuration to your claude_desktop_config.json file according to the methods that suit your needs.

Using pip

(Option 1) Install the released version from PyPI

pip install chronulus-mcp

(Option 2) Install from Github

git clone https://github.com/ChronulusAI/chronulus-mcp.git
cd chronulus-mcp
pip install .

Add the following content to your JSON configuration file:

{
  "mcpServers": {
    "chronulus-agents": {
      "command": "python",
      "args": ["-m", "chronulus_mcp"],
      "env": {
        "CHRONULUS_API_KEY": "<YOUR_CHRONULUS_API_KEY>"
      }
    }
  }
}

⚠️ Important Note

If you encounter an error like "MCP chronulus-agents: spawn python ENOENT", you may need to provide the full Python path, e.g., /Library/Frameworks/Python.framework/Versions/3.11/bin/python3 instead of just python.

Using Docker

(Option 1) Build the Docker image from the source

docker build -t chronulus-mcp .

Using a Third - Party Server

For example, add the following content to your JSON configuration file:

{
  "filesystem": {
    "command": "uvx",
    "args": ["mcp-server-filesystem", "/path/to/AIWorkspace"]
  },
  "fetch": {
    "command": "uvx",
    "args": ["mcp-server-fetch"]
  }
}

⚙️ Configuration Preferences

To optimize your experience of using Claude across multiple toolkits, it is recommended to add the following content to your Claude settings:

From the Claude Desktop: Settings -> General -> Claude Settings -> User Profile (Tab)
From claude.ai/settings: User Profile (Tab)

Here is an example of the preferences used to achieve the demonstration results:

## Tool Dependency Protocol
The following instructions only apply when the tool/MCP server is available.

### File System - Tool Instructions
- Do not use 'read_file' or 'read_multiple_files' on binary files (e.g., images, PDFs, docx).
- When dealing with binary files (e.g., images, PDFs, docx), use the 'get_info' tool instead of 'read_*' tools to check the files.

### Chronulus Agents - Tool Instructions
- When using Chronulus, prefer input field types such as TextFromFile, PdfFromFile, and ImageFromFile instead of directly scanning files.
- When plotting prediction charts from Chronulus, always include the prediction explanation provided by Chronulus below the chart and label it as a Chronulus explanation.