# RoostGPT AI Keys Requirements

# AI Configuration Keys

This document provides the required environment variables for configuring different AI providers.

---

## 1. OpenAI Configuration

### Required Environment Variables

<table id="bkmrk-variable-name-requir" style="width: 73.6905%;"><thead><tr><th style="width: 22.3301%;">Variable Name</th><th style="width: 12.2977%;">Required</th><th style="width: 65.3722%;">Description</th></tr></thead><tbody><tr><td style="width: 22.3301%;">`OPENAI_API_KEY`</td><td style="width: 12.2977%;">Yes</td><td style="width: 65.3722%;">Your OpenAI API key</td></tr><tr><td style="width: 22.3301%;">`OPENAI_API_MODEL`</td><td style="width: 12.2977%;">Yes</td><td style="width: 65.3722%;">Model to use </td></tr></tbody></table>

### Configuration Example

```bash
OPENAI_API_KEY=sk-proj-xxxxxxxxxxxxxxxxxxxx
OPENAI_API_MODEL=gpt-4o
```

---

## 2. Google Gemini Configuration

### Required Environment Variables

<table id="bkmrk-variable-name-requir-1"><thead><tr><th>Variable Name</th><th>Required</th><th>Description</th></tr></thead><tbody><tr><td>`GEMINI_API_KEY`</td><td>Yes</td><td>Your Google AI Studio API key for accessing Gemini models</td></tr><tr><td>`GEMINI_MODEL`</td><td>Yes</td><td>Gemini model to use. Options: `gemini-pro`, `gemini-pro-vision`, `gemini-ultra`, etc.</td></tr></tbody></table>

### Configuration Example

```bash
GEMINI_API_KEY=AIzaSyxxxxxxxxxxxxxxxxxxxxxxxxxx
GEMINI_MODEL=gemini-pro
```

---

## 3. AWS Bedrock Configuration

### Required Environment Variables

<table id="bkmrk-variable-name-requir-2"><thead><tr><th>Variable Name</th><th>Required</th><th>Description</th></tr></thead><tbody><tr><td>`AWS_BEDROCK_MODEL`</td><td>Yes</td><td>Model ID to use. Examples: `anthropic.claude-v2`, `anthropic.claude-3-sonnet-20240229-v1:0`, `amazon.titan-text-express-v1`</td></tr><tr><td>`AWS_DEFAULT_REGION`</td><td>Yes</td><td>AWS region where Bedrock is available. Examples: `us-east-1`, `us-west-2`, `eu-west-1`</td></tr><tr><td>`AWS_ACCESS_KEY_ID`</td><td>Yes</td><td>AWS access key ID for authentication</td></tr><tr><td>`AWS_SECRET_ACCESS_KEY`</td><td>Yes</td><td>AWS secret access key for authentication</td></tr></tbody></table>

### Configuration Example

```bash
AWS_BEDROCK_MODEL=anthropic.claude-3-sonnet-20240229-v1:0
AWS_DEFAULT_REGION=us-east-1
AWS_ACCESS_KEY_ID=AKIAIOSFODNN7EXAMPLE
AWS_SECRET_ACCESS_KEY=wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY
```

### Important Notes

- Model availability varies by region. Check AWS Bedrock documentation for supported models in your region.
- You may need to request model access through the AWS Bedrock console before using certain models.
- Ensure your IAM user has necessary permissions for `bedrock:InvokeModel` action.

---

## 4. Claude AI Configuration

### Required Environment Variables

<table id="bkmrk-variable-name-requir-3"><thead><tr><th>Variable Name</th><th>Required</th><th>Description</th></tr></thead><tbody><tr><td>`CLAUDE_AI_API_KEY`</td><td>Yes</td><td>Your Anthropic API key for accessing Claude models</td></tr><tr><td>`CLAUDE_AI_MODEL`</td><td>Yes</td><td>Claude model to use. Examples: `claude-3-opus-20240229`, `claude-3-sonnet-20240229`, `claude-3-haiku-20240307`, `claude-2.1`, `claude-2.0`</td></tr></tbody></table>

### Configuration Example

```bash
CLAUDE_AI_API_KEY=sk-ant-api03-xxxxxxxxxxxxxxxxxxxxxxxxxx
CLAUDE_AI_MODEL=claude-3-sonnet-20240229
```

---

## 5. Azure OpenAI Configuration

### Required Environment Variables

<table id="bkmrk-variable-name-requir-4"><thead><tr><th>Variable Name</th><th>Required</th><th>Description</th></tr></thead><tbody><tr><td>`AZURE_OPENAI_ENDPOINT`</td><td>Yes</td><td>Your Azure OpenAI resource endpoint URL (e.g., `https://your-resource.openai.azure.com/`)</td></tr><tr><td>`AZURE_DEPLOYMENT_NAME`</td><td>Yes</td><td>Name of your deployed model in Azure OpenAI Studio (e.g., `gpt-4-deployment`, `gpt-35-turbo-deployment`)</td></tr><tr><td>`AZURE_OPENAI_KEY`</td><td>Yes</td><td>API key for your Azure OpenAI resource (Key 1 or Key 2 from Azure portal)</td></tr></tbody></table>

### Configuration Example

```bash
AZURE_OPENAI_ENDPOINT=https://your-resource-name.openai.azure.com/
AZURE_DEPLOYMENT_NAME=gpt-4-deployment
AZURE_OPENAI_KEY=1234567890abcdef1234567890abcdef
```

### Important Notes

- Azure OpenAI requires approval. Apply for access if you haven't already.
- Model availability varies by region. Choose your region based on available models.
- Deployment names are custom - you choose them when deploying models in Azure OpenAI Studio.
- The endpoint URL should end with a trailing slash (`/`).