skills.rest
lanceyuu
lanceyuuCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

deep-research

Runs autonomous Gemini research tasks that plan, search, and synthesize cited reports.

2|1|Updated Aug 19, 2026
Search Tags:#report generation#market analysis#literature review#gemini#competitive analysis#deep research#autonomous research
One-click install
npx skills add https://github.com/lanceyuu/mimiwork --skill deep-research-lanceyuu
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/lanceyuu/mimiwork/tree/main/coworker/skills/builtin/deep-research
Command: npx skills add https://github.com/lanceyuu/mimiwork --skill deep-research-lanceyuu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires httpx, and includes scripts (resource) components.

What problem does it solve? Manual research across many sources is slow and hard to consolidate. This Skill delegates multi-step research to the Gemini Deep Research API, which plans, searches, reads, and synthesizes information into a comprehensive cited report without manual browsing. ## Core Features & Use Cases - Autonomous Research Execution: Submit a query and let the agent plan, search, read, and synthesize findings into a structured markdown report. - Progress Monitoring & Continuation: Stream progress in real time, poll status by interaction ID, and continue prior research with follow-up questions. - Structured Output Formats: Request custom report outlines, JSON output for programmatic use, or raw API responses. - Use Case: Ask for a competitive landscape of EV battery manufacturers; the Skill runs 2-10 minutes of autonomous research and returns a cited report you can refine with follow-up queries. ## Quick Start Set your GEMINI_API_KEY environment variable, then ask the assistant to research a topic such as the history of Kubernetes using the deep-research skill.

Frequently Asked Questions about deep-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run an autonomous deep research task with Gemini?▼

Run scripts/research.py with the --query flag followed by your research question, after setting the GEMINI_API_KEY environment variable. The task plans, searches, reads, and synthesizes sources into a markdown report in about 2-10 minutes.

How do I monitor a long-running Gemini research task?▼

Use --stream to watch progress in real time, or start with --no-wait and poll later using --status with the interaction ID. You can also block until completion with --wait followed by the interaction ID.

Can I continue or refine a previous Gemini research session?▼

Yes, pass --continue with the previous interaction ID along with a new --query to ask follow-up questions. Use --list to see recent research interactions and find the ID you need.

What output formats does the deep research script support?▼

The default output is a human-readable markdown report. Use --json for structured data suitable for programmatic processing, --raw for the unprocessed API response, or --format to specify a custom report outline.

How much does a Gemini deep research task cost and how long does it take?▼

Each task typically takes 2-10 minutes and costs roughly $2-5 depending on complexity. Token usage is approximately 250k-900k input tokens and 60k-80k output tokens per task.

Why does the research script fail with a configuration error?▼

Failures usually come from a missing GEMINI_API_KEY environment variable or missing httpx dependency. Set the key via export or a .env file in the skill directory, and install requirements with pip install -r requirements.txt.

Related Skills

Explore More Education & Research Skills→
K-Dense-AIK-Dense-AI
282

surveying-prior-work

Compile established methods and prior effect sizes to ground analysis planning.

Official
Advanced
MuzeWinterMuzeWinter
1

report-to-google-slides

Convert HTML analytics reports into editable PPTX decks for Google Slides.

Community
Advanced
SigmacodeatSigmacodeat

academic-verify

Trace academic claims to source publications and assess citation accuracy.

Community
Advanced
tawf-labstawf-labs
1

tavily-research

Conduct multi-source web research and synthesize cited reports.

Official
Intermediate