AI Connections
Configure AI provider API keys for your agents
AI Connections lets you configure the LLM provider API keys your agents use. Each agent manages its own provider configurations through the AI Providers tab in Agent Studio.
Configuring a Provider
- Open your agent in Agent Studio
- Open the Configuration Panel (left sidebar)
- Select the AI Providers tab
- Find the provider you want to configure (e.g., OpenAI, Anthropic, Google)
- Expand the provider card
- Enter your API key and any required credentials
- Click Test to validate the connection
- Click Save
Each provider card shows a Configured badge once your API key is saved and validated, or Setup Needed if credentials are missing.
Supported Providers
Fruxon supports multiple AI providers. Each provider has its own set of configuration fields and available models:
- OpenAI — GPT-4o, GPT-4 Turbo, GPT-4, GPT-3.5 Turbo, and more
- Anthropic — Claude 4 Sonnet, Claude 3.5 Sonnet, Claude 3.5 Haiku, and more
- Google — Gemini 2.0 Flash, Gemini 1.5 Pro, Gemini 1.5 Flash, and more
Configuration fields vary by provider but typically include:
| Field | Description |
|---|---|
| API Key | Your provider's API key (stored securely) |
| Additional credentials | Provider-specific fields (e.g., project ID, region) |
Click the Get API key link on any provider card to open the provider's website where you can generate your key.
Using Providers in Workflows
Once configured, providers become available when adding Agent Steps to your workflow:
- Add or select an Agent Step on the canvas
- In the step configuration, select your AI Provider
- Choose a Model from the available options
- Providers with saved API keys show a Configured badge in the dropdown
Agent-Level Configuration
AI provider keys are configured per agent — each agent can use different API keys or different providers. This allows you to:
- Use separate API keys for different agents
- Test with one provider and deploy with another
- Manage billing across different projects or teams
Next Steps
- Agent Studio - Build workflows with AI providers
- Cost & Pricing - Track LLM costs and usage
- Creating Agents - Build your first agent