What is Gemini Deep Research?

An agent that performs complex, long-running context gathering and synthesis tasks, breaking queries into sub-questions, searching systematically, and delivering comprehensive reports with progress updates.

What are the prerequisites?

GEMINI_API_KEY environment variable; does not work with Antigravity OAuth tokens.

Where are results saved?

Results are saved to timestamped files named deep-research-YYYY-MM-DD-HH-MM-SS.md (final report) and deep-research-YYYY-MM-DD-HH-MM-SS.json (full interaction metadata).

Gemini Deep Research

Verified

@arun-8687

npx machina-cli add skill @arun-8687/gemini-deep-research --openclaw

Files (1)

SKILL.md

2.2 KB

Gemini Deep Research

Use Gemini's Deep Research Agent to perform complex, long-running context gathering and synthesis tasks.

Prerequisites

GEMINI_API_KEY environment variable (from Google AI Studio)
Note: This does NOT work with Antigravity OAuth tokens. Requires a direct Gemini API key.

How It Works

Deep Research is an agent that:

Breaks down complex queries into sub-questions
Searches the web systematically
Synthesizes findings into comprehensive reports
Provides streaming progress updates

Usage

Basic Research

scripts/deep_research.py --query "Research the history of Google TPUs"

Custom Output Format

scripts/deep_research.py --query "Research the competitive landscape of EV batteries" \
  --format "1. Executive Summary\n2. Key Players (include data table)\n3. Supply Chain Risks"

With File Search (optional)

scripts/deep_research.py --query "Compare our 2025 fiscal year report against current public web news" \
  --file-search-store "fileSearchStores/my-store-name"

Stream Progress

scripts/deep_research.py --query "Your research topic" --stream

Output

The script saves results to timestamped files:

deep-research-YYYY-MM-DD-HH-MM-SS.md - Final report in markdown
deep-research-YYYY-MM-DD-HH-MM-SS.json - Full interaction metadata

API Details

Endpoint: https://generativelanguage.googleapis.com/v1beta/interactions
Agent: deep-research-pro-preview-12-2025
Auth: x-goog-api-key header (NOT OAuth Bearer token)

Limitations

Requires Gemini API key (get from Google AI Studio)
Does NOT work with Antigravity OAuth authentication
Long-running tasks (minutes to hours depending on complexity)
May incur API costs depending on your quota

Source

git clone https://clawhub.ai/arun-8687/gemini-deep-researchView on GitHub

Overview

Gemini Deep Research automates complex, long-running context gathering and synthesis tasks using Gemini's Deep Research Agent. It is ideal for topics requiring multi-source synthesis, competitive analysis, market research, or comprehensive technical investigations. The agent breaks queries into sub-questions, searches systematically, and delivers thorough reports with streaming progress updates.

How This Skill Works

The Deep Research agent decomposes complex queries into sub-questions, performs systematic web searches, and synthesizes findings into comprehensive reports. It provides streaming progress updates and saves results to timestamped files (deep-research-YYYY-MM-DD-HH-MM-SS.md and .json) via the Gemini API endpoint (https://generativelanguage.googleapis.com/v1beta/interactions) using the x-goog-api-key authentication and the deep-research-pro-preview-12-2025 agent.

When to Use It

Competitive landscape analysis (e.g., EV batteries) to identify key players and market dynamics
Market research and sizing for a new product or category in a target region
Comprehensive technical investigations requiring diverse sources and structured outputs
Vendor or technology stack benchmarking against peers with data-backed findings
Historical or trend research (e.g., development timelines of AI accelerators like Google TPUs)

Quick Start

Step 1: Ensure GEMINI_API_KEY is set in your environment
Step 2: Run a query with scripts/deep_research.py --query "Your topic"
Step 3: Optionally add --format, --stream, or --file-search-store and review the generated deep-research-*.md and .json files

Best Practices

Define sub-questions and the desired output format before starting the research
Use --stream for ongoing progress and --format to enforce structured deliverables
Leverage --file-search-store to incorporate internal documents or datasets
Monitor API usage and potential costs; scope the task to stay within quotas
Cross-check findings with primary sources and clearly cite sources in the final report

Example Use Cases

Research the competitive landscape of EV batteries
Compare our 2025 fiscal-year report with current public news
Benchmark AI accelerator vendors and compute market shares
Map regulatory and standards changes across regions affecting our product
Historical analysis of Google TPUs development timeline

Frequently Asked Questions

Add this skill to your agents