GPT Researcher Development Skill

GPT Researcher is an LLM-based autonomous agent using a planner-executor-publisher pattern with parallelized agent work for speed and reliability.

Instruction

• Initialize the GPTResearcher agent by specifying the research query, report type (e.g., deep, outline), and source (web, local).
• Configure retrieval systems by registering and selecting appropriate search engines in gpt_researcher/retrievers/.
• Define the research plan by decomposing the main query into focused sub-queries for parallel execution.
• Orchestrate the "Planner-Executor-Publisher" pattern to gather context, analyze data, and synthesize final Markdown reports.
• Integrate specialized features like image generation or MCP (Model Context Protocol) data sources into the research workflow.
• Implement error handling with graceful degradation, ensuring the agent remains functional during network or API failures.
• Utilize WebSocket streaming to provide real-time status updates and incremental report generation to the user.

When to Use

• When building autonomous research agents for high-speed, parallelized data gathering and reporting.
• When performing deep-dive research into complex topics requiring synthesis of information from multiple web or local sources.
• When automating the end-to-end research-to-reporting pipeline with high reliability and parallelization.

Output

• Synthesized Markdown research reports containing citations and structured analysis.
• Real-time research logs and status updates delivered via WebSocket or console.
• Architectural diagrams and configuration templates for custom research agent development.