utils/gemini

Scan large codebases with Gemini Flash and generate structured summaries.

Updated Jan 25, 2026
One-click install
npx skills add https://github.com/codihaus/claude-skills --skill utils-gemini
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: utils/gemini
Source: https://github.com/codihaus/claude-skills/tree/main/skills/utils/gemini
Command: npx skills add https://github.com/codihaus/claude-skills --skill utils-gemini

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-generativeai, and includes scripts (resource) components.

What problem does it solve?

In contexts where Claude's context window is exceeded by a large codebase, Gemini Flash provides a scalable initial scan and summarization to surface key insights.

Core Features & Use Cases

  • Large-context scanning for massive codebases to surface actionable summaries
  • Structured summaries that Claude can ingest and refine into scout outputs
  • Quick handoff for large projects to drive deeper analysis with minimal context switching

Quick Start

Run the Gemini scan on your project to generate a structured summary that Claude can use to bootstrap a scout

Frequently Asked Questions about utils/gemini

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

FAQPage Schema
How do I analyze a large codebase that exceeds Claude's context window?

To analyze a large codebase exceeding Claude's context window, this Skill uses Gemini Flash to scan dense repositories with 500+ files, producing a structured summary that Claude can ingest for deeper analysis.

What is large-context codebase scanning and how does it surface codebase insights?

Large-context codebase scanning uses models like Gemini Flash to read massive project files beyond standard limits, generating structured summaries that surface actionable insights for subsequent AI workflows.

Do I need a Gemini API key to run large-context codebase analysis?

Yes, you need a Gemini API key to run large-context codebase analysis, as the Skill relies on the google-generativeai dependency to execute the scan and generate the summary output.

How do I configure ignore rules and output formats for codebase scanning?

You configure deterministic behavior for codebase scanning by setting up configurable extensions and ignore rules, ensuring the Gemini-derived summary matches your specific output format requirements.

When should I use Gemini Flash instead of Claude for codebase analysis?

Use Gemini Flash instead of Claude for codebase analysis when targeting projects with 500+ files or dense repositories that exceed Claude's context window, allowing a quick handoff for deeper scout workflows.