Personal Productivity Agent

What is Personal Productivity Agent?

Personal Productivity Agent is a Python-based tool for Windows systems that helps with file operations, system information, and productivity tasks, and uses Google’s Gemini LLM to analyze Windows Event Logs. It can run as an interactive command-line assistant or as an MCP server for programmatic integration.

How to use Personal Productivity Agent?

Ensure Python 3.x is installed, set the GEMINI_API_KEY environment variable, and install the required package (google-generativeai). Run python main_agent.py for the interactive CLI, or python main_agent.py --mcp-mode to start the MCP server that accepts JSON requests on stdin.

Key features of Personal Productivity Agent

• Find, move, rename files and create directories (with confirmation)
• Get disk usage, memory, CPU, and network information
• Launch applications and find duplicate files
• Summarize documents (currently using a mock LLM)
• Retrieve upcoming events from .ics calendar files
• Set desktop reminders/notifications
• Analyze Windows Event Logs with Gemini LLM (suggested causes and fixes)

Use cases of Personal Productivity Agent

• Automate file organization and cleanup tasks on a Windows system
• Monitor system health (disk, memory, network) from the command line
• Analyze Windows Event Logs with LLM explanations and troubleshooting steps
• Integrate the agent as a backend tool for a larger orchestration system via MCP
• Schedule or trigger desktop reminders based on calendar events

FAQ from Personal Productivity Agent

What LLM does the agent use?

The agent uses Google’s Gemini LLM for event log analysis. A GEMINI_API_KEY environment variable is required; if not set, a mock LLM is used.

What are the system requirements?

Python 3.x is needed, and the google-generativeai package must be installed. The agent is designed for Windows (uses Windows Event Logs, file system operations, and notifications).

How do I run the agent in MCP server mode?

Run python main_agent.py --mcp-mode. The agent then listens for JSON-formatted requests on stdin and returns JSON responses on stdout.

Can the agent be integrated with other applications?

Yes, in MCP mode it communicates via a simple JSON protocol over stdio, making it easy to connect with orchestrators or TypeScript-based agents.

How do I set up the Gemini API key?

Set the environment variable GEMINI_API_KEY in PowerShell (e.g., $env:GEMINI_API_KEY = "YOUR_API_KEY_HERE") before launching the agent. Do not commit the key to version control.