Neura Market
deep-research logo

deep-research

Free

Execute autonomous multi-step research using Gemini Deep Research Agent for market analysis, competitive landscaping, and literature reviews. *By [@sanjay3290](https://github.com/sanjay3290)*

FreeFree tier
Inputs: text
Type
Open Source
Visit Website GitHub (40 stars)

About deep-research

deep-research is an open-source skill that enables autonomous multi-step research tasks using Google's Gemini Deep Research Agent (deep-research-pro-preview-12-2025). It functions as an 'analyst-in-a-box' that plans research strategy, searches the web, analyzes sources, reads and extracts relevant information, iterates through multiple search/read cycles, and outputs detailed, cited reports. The process takes 2–10 minutes and produces thorough analysis equivalent to hours of human research. It supports real-time streaming of progress, structured JSON output, follow-up conversations, and configurable timeouts.

Key Features

Autonomous multi-step research: plans, searches, reads, iterates, and synthesizes information
Powered by Gemini 3 Pro (deep-research-pro-preview-12-2025)
Real-time streaming of research progress and thinking steps
Structured JSON output and raw API response options
Follow-up conversation support to continue previous research
Configurable timeout and poll interval via environment variables
Local history of recent research tasks
Custom output format specification (e.g., executive summary, comparison table, pros/cons)

Pros & Cons

Pros
  • Produces thorough, cited reports that would take hours of human effort
  • Autonomous end-to-end research with minimal user input
  • Real-time streaming allows monitoring progress as it happens
  • Flexible output formats (markdown, JSON, raw API) suit different needs
  • Supports follow-up questions for iterative exploration
  • Open-source and free to use (requires own Gemini API key)
Cons
  • Requires a Google Gemini API key (may incur usage costs)
  • Research takes 2–10 minutes, slower than standard LLM queries
  • Requires local installation and Python environment setup
  • Depends on external web search and API availability
  • No built-in GUI; command-line interface only

Best For

Market analysis and competitive landscapingLiterature reviews and historical researchTechnology comparison (e.g., cloud providers, frontend frameworks, EV market)Generating structured reports with citationsFollow-up deep dives on specific sections of previous research

FAQ

How do I get a Gemini API key?
Go to Google AI Studio, click 'Get API key', and create a new key or use an existing one. Then copy the key into the .env file.
How long does a research task take?
The process takes 2–10 minutes but produces thorough analysis that would take a human researcher hours.
Can I stream progress in real-time?
Yes, use the --stream flag to see research progress and thinking steps as they are generated.
Can I continue a previous research conversation?
Yes, use the --continue flag with the interaction ID to ask follow-up questions on earlier research.
Is the tool free to use?
The skill itself is open-source and free, but you need a Gemini API key which may have usage costs depending on your Google Cloud plan.