From more-ai
Consults Google's Gemini model for second opinions on coding tasks like debugging tricky issues, evaluating architecture trade-offs, reviewing code, and brainstorming alternatives.
npx claudepluginhub nityeshaga/claude-home-base --plugin more-aiThis skill uses the workspace's default tool permissions.
Consult Google's Gemini model for a second opinion on any task. Run the script, pass your question, and incorporate the response into your own thinking.
Invokes OpenAI Codex and Google Gemini CLIs for adversarial code reviews, tie-breaking, and multi-model consensus on critical decisions like security and architecture.
Queries AI models via OpenRouter, Gemini, or OpenAI APIs for second opinions on code, architecture, strategy, or prompting. Supports consensus, single opinion, and devil's advocate modes.
Invokes OpenAI Codex and Google Gemini CLIs via Bash for second opinions, code reviews, and alternative analysis. Useful when users request external AI verification or explicitly say 'ask codex' or 'ask gemini'.
Share bugs, ideas, or general feedback.
Consult Google's Gemini model for a second opinion on any task. Run the script, pass your question, and incorporate the response into your own thinking.
Check if GEMINI_API_KEY is set in the environment. If not, ask the user to configure it:
~/.zshrc: export GEMINI_API_KEY=your_key_heresource ~/.zshrcpython3 scripts/gemini_think.py "Your question here" [--model MODEL] [--file FILE] [--system INSTRUCTION] [--thinking-budget N]
| Option | Values |
|---|---|
--model | gemini-3-pro-preview (default, strongest reasoning), gemini-3-flash-preview (fast, balanced) |
--file | Path to a file to include as context |
--system | System instruction to guide the model's behavior |
--thinking-budget | Max thinking tokens, 0-8192 (default: 8192 = maximum reasoning depth) |
Use this as a thinking tool, not a replacement for your own judgment. Good patterns:
--file and ask for a review.# Quick question (uses Gemini 3 Pro with max thinking by default)
python3 scripts/gemini_think.py "What are the trade-offs between REST and GraphQL for a read-heavy internal API?"
# Code review with file context
python3 scripts/gemini_think.py --file src/auth.py "Review this auth module for security issues"
# Fast model for simple questions
python3 scripts/gemini_think.py --model gemini-3-flash-preview "Is it better to use a Set or Array for deduplication in JavaScript?"
# Lower thinking budget for quick answers
python3 scripts/gemini_think.py --thinking-budget 1024 "What's the difference between a mutex and a semaphore?"
# With system instruction for focused feedback
python3 scripts/gemini_think.py --system "You are a senior security engineer" --file config.yaml "Audit this configuration for security concerns"
gemini-3-pro-preview for complex reasoning. Use gemini-3-flash-preview for quick factual questions.--thinking-budget 1024 for faster, cheaper responses.--file or inline in the prompt.