Skip to main content
Generalcaidish

gemini-cli

Wield Google's Gemini CLI as a powerful auxiliary tool for code generation, review, analysis, and web research. Use when tasks benefit from a second AI perspective, current web information via Google Search, codebase architecture analysis, or parallel code generation. Also use when user explicitly requests Gemini operations.

Stars
40
Source
caidish/cAI-tools
Updated
2026-05-28
Slug
caidish--cAI-tools--gemini-cli
View on GitHubRaw SKILL.md

// install — copy + paste into any project

mkdir -p .claude/skills && curl -fsSL https://raw.githubusercontent.com/caidish/cAI-tools/HEAD/plugins/AI-skill/skills/gemini-cli/SKILL.md -o .claude/skills/gemini-cli.md

Drops the SKILL.md into .claude/skills/gemini-cli.md. Works with Claude Code, Cursor, and any agent that loads SKILL.md files from .claude/skills/.

Gemini CLI Skill Guide

When to Use Gemini

Use Case Why Gemini
Current web information google_web_search - real-time Google Search
Codebase architecture analysis codebase_investigator - deep analysis tool
Second opinion / code review Different AI perspective catches different bugs
Parallel code generation Offload tasks while continuing other work

When NOT to use: Simple quick tasks (overhead not worth it), interactive refinement, context already understood.

Running a Task

  1. Verify installation: command -v gemini
  2. Select the mode required for the task; default to read-only (no --yolo) unless edits are necessary.
  3. Always use AskUserQuestion before using --yolo or -s flags. These modes allow file writes or sandboxed execution - get explicit user approval first.
  4. Assemble the command with appropriate options:
    • -m, --model <MODEL> - Model selection
    • -y, --yolo - Auto-approve all tool calls (enables writes)
    • -s, --sandbox - Run in Docker isolation
    • -o, --output-format <text|json> - Output format
  5. Important: gemini "prompt" -o json 2>/dev/null | jq -r '.response' to suppress stderr noise and extract the json response, unless specified by the user.

Critical Note

YOLO mode does NOT prevent planning prompts. Use forceful language: "Apply now", "Start immediately", "Do this without asking for confirmation".

Quick Reference

Use case Mode Command pattern
Read-only analysis read-only gemini "..." -o json 2>/dev/null | jq -r '.response'
Apply local edits write gemini "..." --yolo -o json 2>/dev/null | jq -r '.response'
Sandboxed write sandbox gemini "..." --yolo --sandbox -o json 2>/dev/null | jq -r '.response'

Example Commands

# Read-only
gemini "Review src/ for bugs" -o json 2>/dev/null | jq -r '.response'

# Write mode
gemini "Fix bug in file.py. Apply now." --yolo -o json 2>/dev/null | jq -r '.response'

# If redirection fails, wrap in bash -lc
bash -lc 'gemini "prompt" -o json 2>/dev/null | jq -r ".response"'

Following Up

  • Resume: echo "follow-up" | gemini -r latest -o json 2>/dev/null | jq -r '.response'
  • List sessions: gemini --list-sessions

Error Handling

  • Rate limit: CLI auto-retries with backoff. Use -m gemini-2.5-flash for lower priority tasks.
  • Command failure: Check with gemini --version, use --debug for details.
  • Always validate Gemini's output for security vulnerabilities (XSS, injection) before using.