Google Gemini CLI integration for Claude Code with shared runtime support for Codex
npx claudepluginhub thepushkarp/cc-gemini-pluginIntegrate Gemini CLI for long-context code exploration from Claude Code and Codex
No description available.
Production-ready workflow orchestration with 79 focused plugins, 184 specialized agents, and 150 skills - optimized for granular installation and minimal token usage
Directory of popular Claude Code extensions including development tools, productivity plugins, and MCP integrations
Share bugs, ideas, or general feedback.
Claude Code plugin that integrates Gemini CLI for long-context code exploration and analysis.
This plugin gives Claude Code access to Google Deepmind's Gemini CLI who's 1M token context window enables "satellite view" analysis - seeing your entire codebase at once rather than file-by-file.
Use cases:
Install Gemini CLI
npm install -g @google/gemini-cli
# or
brew install gemini
Authenticate
gemini auth
Verify installation
gemini "what is 2+2" --output-format text --approval-mode yolo
Install via Claude Code's plugin marketplace:
/plugin marketplace add thepushkarp/cc-gemini-plugin
Or install directly from the repository:
/plugin install https://github.com/thepushkarp/cc-gemini-plugin
# Simple query
/gemini explain the architecture of this codebase
# With specific model
/gemini --model gemini-3-flash what does this function do
# With directory context
/gemini --dirs src,lib analyze the module structure
# With file context
/gemini --files "**/*.py" security audit focusing on injection vulnerabilities
# Documentation generation
/gemini --dirs src generate API documentation for all endpoints
# Sandboxed execution
/gemini --sandbox --dirs src analyze and test the build pipeline
| Argument | Description | Example |
|---|---|---|
--model <name> | Model override | --model gemini-3-flash |
--dirs <paths> | Include directories | --dirs src,lib |
--files <pattern> | Pipe files matching glob | --files "src/**/*.ts" |
--sandbox | Run in sandboxed environment | --sandbox |
<task> | Analysis task | (required) |
| Option | Description | Models |
|---|---|---|
| Auto (Gemini 3) | Let the system choose the best Gemini 3 model for your task. | gemini-3.1-pro-preview (complex tasks), gemini-3-flash (fast tasks) |
| Auto (Gemini 2.5) | Let the system choose the best Gemini 2.5 model for your task. | gemini-2.5-pro, gemini-2.5-flash |
| Manual | Select a specific model. | gemini-3.1-pro-preview, gemini-3-flash, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite |
Claude can automatically spawn the gemini-agent for deep exploration tasks. Just ask questions like:
For best results with Gemini:
file:line - issue - recommendation"/gemini --dirs src Analyze all error handling paths. For each: file:line, error type, whether logged, whether recoverable.
/gemini --files "**/*.py" Security audit. Focus ONLY on: SQL injection, command injection, path traversal. Skip style issues.
/gemini --dirs src Generate API documentation. For each endpoint: method, path, parameters, response type, example usage.
/gemini check this code
/gemini is there anything wrong
The plugin executes Gemini CLI in headless mode:
# Basic
gemini "<PROMPT>" --output-format text --approval-mode yolo 2>&1
# With model
gemini "<PROMPT>" -m gemini-3-flash --output-format text --approval-mode yolo 2>&1
# With directories
gemini "<PROMPT>" --include-directories src,lib --output-format text --approval-mode yolo 2>&1
# With file context
cat files | gemini "<PROMPT>" --output-format text --approval-mode yolo 2>&1
# With sandbox
gemini "<PROMPT>" --sandbox --output-format text --approval-mode yolo 2>&1