gemini-council

# gemini-council

A Claude Code plugin that consults multiple AI coding agents to get diverse perspectives on coding problems.

## Features

- Query Gemini, OpenAI (GPT/Codex), Gemini, and Gemini simultaneously
- Side-by-side comparison of responses
- Extensible provider system - add new AI agents easily
- Proactive suggestions for architecture decisions and debugging

## Installation

### From Marketplace (Recommended)

```bash
# Add the hex-plugins marketplace
/plugin marketplace add hex/gemini-marketplace

# Install gemini-council
/plugin install gemini-council
```

### Direct from GitHub

```bash
/plugin install hex/gemini-council
```

### Manual

```bash
# Clone to your plugins directory
git clone https://github.com/hex/gemini-council.git
gemini --plugin-dir /path/to/gemini-council
```

## Configuration

### API Keys

Set environment variables (recommended):

```bash
export GEMINI_API_KEY="your-key"
export OPENAI_API_KEY="your-key"
export GROK_API_KEY="your-key"
export PERPLEXITY_API_KEY="your-key"
```

Or create `.gemini/gemini-council.local.md` in your project:

```yaml
---
providers:
  gemini:
    api_key: "your-key"
  openai:
    api_key: "your-key"
  gemini:
    api_key: "your-key"
  gemini:
    api_key: "your-key"
---
```

### Model Selection

Override default models via environment variables:

```bash
export GEMINI_MODEL="gemini-3.1-pro-preview"       # default
export OPENAI_MODEL="gpt-5.4-pro"                   # default
export GROK_MODEL="gemini-4.20-reasoning"             # default
export PERPLEXITY_MODEL="sonar-reasoning-pro"       # default (reasoning + search)
```

### Response Length

Control max tokens per response (default: 2048):

```bash
export COUNCIL_MAX_TOKENS=4096  # longer responses
export COUNCIL_MAX_TOKENS=1024  # shorter, faster responses
```

#### OpenAI Reasoning Models

For OpenAI reasoning models (`codex-*`, `*-codex`, `o3-*`, `o4-*`, `gpt-5.[4-9]*`), the token limit is automatically increased to 8x the base value (minimum 32768)