Refactor high-complexity React components in Dify frontend. Use when `pnpm analyze-component...
npx skills add sam-mfb/claude-meta --skill "gemini-cli"
Install specific skill from multi-skill repository
# Description
Use the Gemini CLI to invoke Google's Gemini AI from the command line. Use when you need a second AI opinion, want to verify your work, or need Gemini-specific capabilities.
# SKILL.md
name: gemini-cli
description: Use the Gemini CLI to invoke Google's Gemini AI from the command line. Use when you need a second AI opinion, want to verify your work, or need Gemini-specific capabilities.
argument-hint: [prompt]
allowed-tools: Bash, Read
Gemini CLI
Use the gemini command to invoke Google's Gemini AI from the command line.
Prerequisites
The user must have already:
1. Installed Gemini CLI: npm install -g @google/gemini-cli
2. Logged in via Google account: run gemini and select "Login with Google"
3. Enabled preview features: run gemini, then /settings, toggle "Preview Features" to true
Basic Command
Always use the latest preview model with -m and non-interactive mode with -p:
gemini -m gemini-3-flash-preview -p "Your prompt here"
Models
| Model | Use Case |
|---|---|
gemini-3-flash-preview |
Fast, general purpose |
gemini-3-pro-preview |
Complex reasoning tasks |
File References
Use @ syntax to include files in prompts:
gemini -m gemini-3-flash-preview -p "Review @./src/main.js"
gemini -m gemini-3-flash-preview -p "Compare @./a.txt and @./b.txt"
Piped Input
cat file.txt | gemini -m gemini-3-flash-preview -p "Summarize this"
git diff | gemini -m gemini-3-flash-preview -p "Write a commit message"
JSON Output
For programmatic parsing:
gemini -m gemini-3-flash-preview -p "Query" --output-format json | jq -r '.response'
Key Flags
| Flag | Description |
|---|---|
-p "prompt" |
Non-interactive mode (required for scripting) |
-m model |
Specify model |
--output-format json |
Structured JSON output |
--yolo |
Auto-approve tool executions (use carefully) |
--sandbox |
Run in sandboxed environment |
Exit Codes
| Code | Meaning |
|---|---|
| 0 | Success |
| 1 | General error |
| 41 | Authentication error |
| 42 | Invalid input |
When Using This Skill
If $ARGUMENTS is provided, construct and run the appropriate gemini command for that prompt.
Otherwise, use Gemini CLI when you need to:
- Get a second opinion on code or analysis
- Verify your own output
- Leverage Gemini's specific strengths
# Supported AI Coding Agents
This skill is compatible with the SKILL.md standard and works with all major AI coding agents:
Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.