GPT Researcher Skill

This skill allows you to utilize the GPT Researcher Python package as an autonomous research agent.

Setup

1. Dependencies: Requires gpt-researcher and its dependencies.

   pip install gpt-researcher python-dotenv
   

2. Configuration: GPT Researcher needs API keys for an LLM (e.g., OpenAI) and a Search Provider (Tavily is recommended).

   # Prompt user for keys if not set (assuming OpenAI and Tavily)
   if [ -z "$OPENAI_API_KEY" ] || [ -z "$TAVILY_API_KEY" ]; then
       echo "GPT Researcher requires API keys."
       read -p "Enter your OpenAI API key: " OAI_KEY
       read -p "Enter your Tavily API key: " TAVILY_KEY
       echo "OPENAI_API_KEY=$OAI_KEY" >> .env
       echo "TAVILY_API_KEY=$TAVILY_KEY" >> .env
       if [ -f .gitignore ] && ! grep -q ".env" .gitignore; then echo ".env" >> .gitignore; fi
       echo "API keys saved to .env."
   fi
   

Usage

Use the scripts/run_research.py script to initiate a research task.

Command

python3 scripts/run_research.py --query "<query>" [--report-type <type>]

Parameters

• --query (Required): The topic to research.
• --report-type (Optional): Default research_report. Options include: research_report, resource_report, outline_report, deep_research_report.

Example

python3 scripts/run_research.py --query "The future of renewable energy sources" --report-type deep_research_report

Output

The script outputs the final report in Markdown format to stdout. The research process (which can take several minutes) is logged to stderr.