gemini

Analyze codebases and web sources via Gemini MCP, saving cited reports.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/ryugen04/dotfiles --skill gemini-ryugen04
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gemini
Source: https://github.com/ryugen04/dotfiles/tree/main/packages/claude/.claude/skills/gemini
Command: npx skills add https://github.com/ryugen04/dotfiles --skill gemini-ryugen04

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables objective, citation-backed investigations using Gemini MCP for large-scale codebase research or web information gathering, producing concrete reports with explicit sources and verifiable findings.

Core Features & Use Cases

  • Codebase investigation: perform multi-file analyses with citations to files and line numbers.
  • Web search support: retrieve up-to-date information with source URLs and exact quotes.
  • Long log analysis and context optimization: distill large logs into focused summaries with exact findings.

Quick Start

Use Gemini to generate a report for a topic of interest (e.g., a module or API). The report will be saved under .claude/gemini-reports/{topic}-{YYYYMMDD-HHMM}.md with all findings explicitly enumerated and cited.

Frequently Asked Questions about gemini

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

FAQPage Schema
How do I generate citation-backed reports from codebase analysis?

Use Gemini MCP to analyze your codebase and automatically generate reports with explicit citations to files and line numbers. The Skill produces structured findings saved to .claude/gemini-reports/ with all sources verified and enumerated.

Can I use this Skill to analyze large codebases with web search integration?

Yes. The Skill handles large-scale codebase investigations using @ syntax alongside web search to retrieve up-to-date information with source URLs and exact quotes, producing comprehensive reports that combine both data sources.

What's the best way to distill findings from lengthy logs into focused summaries?

This Skill optimizes long log analysis by extracting key findings with exact citations, condensing verbose output into actionable summaries while preserving source references and maintaining full traceability.

Does this work for investigating specific modules or APIs across multiple files?

Yes. Multi-file analysis with citations to specific modules or APIs is a core use case. The Skill performs cross-file investigation and outputs all findings explicitly enumerated with line-number references.

What output format do the reports use and where are they stored?

Reports are saved as Markdown files in .claude/gemini-reports/{topic}-{YYYYMMDD-HHMM}.md format with all findings explicitly enumerated, cited, and verifiable for downstream use or documentation.

When should I use MCP-based investigation over manual codebase review?

Use this Skill when you need exhaustive, citation-backed analysis at scale—particularly for large codebases, long logs requiring context optimization, or web searches demanding current, sourced information with verifiable findings.