Gemini
Gemini provides creative text generation โ ideal for brainstorming content, crafting stories, and exploring diverse writing styles efficiently.
Install on your platform
We auto-selected Claude Code based on this skillโs supported platforms.
Run in terminal (recommended)
claude mcp add davila7-gemini npx -- -y @trustedskills/davila7-gemini
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"davila7-gemini": {
"command": "npx",
"args": [
"-y",
"@trustedskills/davila7-gemini"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
The Gemini skill enables AI agents to perform comprehensive text generation and analysis, particularly suited for code review, plan review, and processing large amounts of context (over 200k tokens). It leverages various Gemini models โ including gemini-3-pro-preview, gemini-3-flash, gemini-2.5-pro, gemini-2.5-flash, and gemini-2.5-flash-lite - to handle complex tasks efficiently. The skill is designed for both interactive and automated workflows, but requires careful configuration of approval modes.
When to use it
- Code Review: Analyzing code across multiple files to identify potential issues.
- Plan Review: Examining architectural plans, technical specifications, or project roadmaps for inconsistencies or areas needing improvement.
- Large Context Processing: Tasks involving extensive documentation sets or entire codebases that exceed typical token limits.
- Multi-file Analysis: Understanding relationships and patterns across numerous files within a project.
Key capabilities
- Supports multiple Gemini models:
gemini-3-pro-preview,gemini-3-flash,gemini-2.5-pro,gemini-2.5-flash, andgemini-2.5-flash-lite. - Offers different approval modes for interactive (
default), automated editing (auto_edit), or fully automated execution (yolo). - Capable of processing contexts exceeding 200,000 tokens.
- Designed for analyzing relationships and patterns across multiple files.
Example prompts
- "Please review this codebase using the
gemini-3-pro-previewmodel with YOLO approval mode." - "Can you analyze these architectural plans and identify any potential structural weaknesses, using the
gemini-2.5-flashmodel?" - "Review these technical specifications for consistency and completeness, selecting the best available Gemini model."
Tips & gotchas
- Approval Mode is Critical: Never use
--approval-mode defaultin non-interactive environments (like automated tool calls) as it will cause the process to hang indefinitely. Use--approval-mode yolofor fully automated tasks or wrap execution with a timeout (timeout 300 gemini ...). - Model Selection: The skill requires you to specify which Gemini model to use in a single prompt. Consider
gemini-3-pro-previewfor complex reasoning and coding, whilegemini-3-flashprioritizes speed. - Hung Process Symptoms: If the process runs for over 20 minutes with no CPU usage or network activity, it may be hung. Use
ps aux | grep gemini | grep -v grepto check andpkill -9 -f "gemini.*gemini-3-pro-preview"to terminate if necessary.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates โ what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
๐ Community
Passed automated security scans.