Marketplace

gemini-research

Use Gemini CLI for research with Google Search grounding and 1M token context

allowed_tools: Bash, Read, Write, TodoWrite, WebFetch, Glob, Grep

$ Installieren

git clone https://github.com/DNYoussef/context-cascade /tmp/context-cascade && cp -r /tmp/context-cascade/skills/platforms/multi-model-extended/gemini-research ~/.claude/skills/context-cascade

// tip: Run this command in your terminal to install the skill

SKILL.md

View on GitHub →

name: gemini-research description: Use Gemini CLI for research with Google Search grounding and 1M token context allowed-tools: Bash, Read, Write, TodoWrite, WebFetch, Glob, Grep

Gemini Research Skill

Purpose

Route research tasks to Gemini CLI when:

Real-time information is needed (Google Search grounding)
Context exceeds Claude's 200k limit (Gemini has 1M)
Need web-grounded factual answers

Unique Capability

What Gemini Does Better:

Google Search grounding for current information
1M token context for massive document analysis
70+ extensions (Figma, Stripe, Shopify, etc.)
Web content analysis with source attribution

When to Use

Perfect For:

Current events, recent documentation
Large codebase analysis (>150k tokens)
Literature reviews with many papers
Real-time API documentation lookup
Market research, competitor analysis

Don't Use When:

Offline/airgapped environments
Complex multi-step reasoning (use Claude)
Code generation requiring iteration (use Codex)

Usage

Basic Research

/gemini-research "What are the latest React 19 best practices?"

With Context Files

/gemini-research "Analyze architecture" --context @src/

Large Document Analysis

/gemini-research "Summarize all papers" --context papers/*.pdf

Command Pattern

bash scripts/multi-model/gemini-research.sh "<query>" "<task_id>" "json"

Memory Integration

Results stored to Memory-MCP:

Key: multi-model/gemini/research/{task_id}
Tags: WHO=gemini-cli, WHY=research

Output Format

{
  "content": "Research findings...",
  "sources": ["url1", "url2"],
  "model": "gemini-2.5-pro",
  "timestamp": "2025-12-28T..."
}

Handoff to Claude

After Gemini research completes:

Results stored in Memory-MCP
Claude agents read from memory key
Use research to inform implementation

// Claude agent reads Gemini research
const research = memory_retrieve("multi-model/gemini/research/{task_id}");
Task("Coder", `Implement using: ${research.content}`, "coder");

Configuration

Retries: 3 attempts on failure
Timeout: 60 seconds per query
Fallback: Claude researcher agent if Gemini unavailable