By thepushkarp
Leverage Gemini AI's long context window via CLI in Claude Code to explore entire codebases from directories or file globs, performing architecture mapping, refactor impact tracing, analysis, documentation generation, and structured data synthesis from multiple files.
npx claudepluginhub thepushkarp/cc-gemini-plugin --plugin cc-gemini-pluginDual-host Gemini CLI integration for Claude Code and Codex.
This repository uses one shared Gemini runtime and two thin host adapters:
/cc-gemini-plugin:gemini and gemini-agent.gemini-integration skill.It gives each host a clean way to hand large, cross-file analysis tasks to Gemini instead of solving everything file-by-file.
scripts/gemini-bridge.js/cc-gemini-plugin:gemini
command, and gemini-agentSKILL.md skill definition and
agents/openai.yamltests/gemini-bridge.test.jsnpm install -g @google/gemini-cli
# or
brew install gemini-cli
gemini auth
gemini -p "what is 2+2" --output-format text
This is a user-level install. Once you add the marketplace and install the plugin, it stays available in new Claude Code sessions on this machine.
Add the marketplace from GitHub, install the plugin, then reload plugins:
/plugin marketplace add thepushkarp/cc-gemini-plugin
/plugin install cc-gemini-plugin@cc-gemini-plugin
/reload-plugins
After installation, use:
/cc-gemini-plugin:gemini <task>
To update the plugin:
/plugin marketplace update cc-gemini-plugin
/reload-plugins
Codex does not need a plugin for this repository. Install it as a user-level skill so it is available in new Codex sessions on this machine across repositories.
Install it by cloning the repository into ~/.agents/skills:
mkdir -p ~/.agents/skills
git clone https://github.com/thepushkarp/cc-gemini-plugin.git \
~/.agents/skills/cc-gemini-plugin
Restart Codex after cloning the skill.
To update it later:
git -C ~/.agents/skills/cc-gemini-plugin pull
After installation, use the bundled skill:
$gemini-integration
Both hosts route through:
node scripts/gemini-bridge.js [options] <task>
Supported options:
--model <name>--dirs <path,...>--files <glob,...>--format <text|json|stream-json>--max-files <n>--max-file-bytes <n>--print-commandThe bridge:
Use:
/cc-gemini-plugin:gemini <task>
/cc-gemini-plugin:gemini --dirs src,docs <task>
/cc-gemini-plugin:gemini --files "schemas/**/*.json,data/**/*.csv" <task>
Use the bundled skill:
$gemini-integration
Or ask Codex to use the Gemini integration for a large-context pass.
Codex-specific skill metadata lives in agents/openai.yaml.
Architecture review:
node scripts/gemini-bridge.js --dirs src,docs \
"Explain the architecture and cite the key files."
Refactor impact:
node scripts/gemini-bridge.js --dirs src \
"Analyze the impact of refactoring the auth module. Include affected files and migration steps."
Structured data review:
node scripts/gemini-bridge.js --files "schemas/**/*.json,data/**/*.csv" \
"Summarize the data contracts and identify breaking changes."
Structured output:
node scripts/gemini-bridge.js --format json --dirs src \
"Summarize the public API surface."
Run the bridge tests:
npm test
cc-gemini-plugin/
├── .claude-plugin/
│ ├── marketplace.json
│ └── plugin.json
├── SKILL.md
├── agents/
│ ├── gemini-agent.md
│ └── openai.yaml
├── commands/
│ └── gemini.md
├── scripts/
│ └── gemini-bridge.js
├── tests/
│ └── gemini-bridge.test.js
└── package.json
| Issue | Solution |
|---|---|
| Authentication error | Run gemini auth |
| Gemini missing on PATH | Install @google/gemini-cli or brew install gemini-cli |
| Token pressure | Narrow the inlined scope with fewer directories or more specific globs |
| Timeout | Reduce the context set and tighten the task |
MIT
Use when the user asks to run Gemini CLI for code review, plan review, or big context (>200k) processing. Ideal for comprehensive analysis requiring large context windows. Uses Gemini 3 Pro by default for state-of-the-art reasoning and coding.
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Google Gemini CLI ecosystem plugin with documentation management, Claude-to-Gemini integration agents, exploration/planning skills, workspace bridging, and meta-skills for configuration, checkpointing, sandboxing, and session management
OpenAI Codex CLI integration with MCP server and skills for plan review, code review, and codebase analysis
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
AI-powered wiki generator for code repositories. Generates comprehensive, Mermaid-rich documentation with dark-mode VitePress sites, onboarding guides, deep research, and source citations. Inspired by OpenDeepWiki and deepwiki-open.