AI Reply
Intelligent, AI-powered responses to form submissions
AI Reply automatically generates personalized, contextual responses to form submissions using AI. Unlike static auto-responders, AI Reply understands the content of each submission and generates unique, helpful replies based on your knowledge base.
Pro Feature
AI Reply is available on the Pro and Advanced plans. Upgrade to Pro to unlock this feature.
How It Works
AI Reply uses your configured AI provider (OpenAI, Anthropic, or Google) to generate responses:
Form Submitted
User submits your form with their message, question, or inquiry.
AI Generates Reply
AI processes the submission with your knowledge base to craft a relevant response.
Response Sent
The AI-generated reply is automatically emailed to the form submitter.
Bring Your Own Key (BYOK)
AI Reply uses a "Bring Your Own Key" model. You provide your own API key from your preferred AI provider, giving you:
Pay only for what you use directly to the AI provider
Use OpenAI (GPT-4), Anthropic (Claude), or Google (Gemini)
Choose from various models based on your needs
Your API key is encrypted at rest using AES-256 encryption
Setting Up AI Reply
Choose Your AI Provider
Select your preferred AI provider: OpenAI, Anthropic, or Google. Each provider offers different models with varying capabilities. See the Supported Models section below for details.
Enter Your API Key
Get an API key from your chosen provider and enter it in the settings. The key is encrypted before being stored.
OpenAI: Get your key from platform.openai.com
Anthropic: Get your key from console.anthropic.com
Google: Get your key from aistudio.google.com
Add Your Knowledge Base
Provide context for the AI to generate relevant responses:
- Text content: Enter information directly (FAQs, product details, policies)
- Files: Upload .txt or .md files with detailed information (up to 25MB total)
Configure Settings
Customize the AI behavior: set rate limits, exclude sensitive fields, and configure fallback options.
Test Configuration
Use the "Test Configuration" button to verify everything works correctly before going live.
Knowledge Base
The knowledge base provides context for the AI to generate accurate, relevant responses. The more detailed your knowledge base, the better the AI replies will be.
What to Include
Business Information
Company name, services, products, pricing, hours, contact methods
FAQs
Common questions and their answers
Policies
Shipping, returns, privacy, terms of service
Response Guidelines
Tone of voice, key phrases, things to avoid
Knowledge Base Example
Excluding Sensitive Fields
You can exclude certain form fields from being sent to the AI for privacy reasons. Common fields to exclude:
- Credit card numbers
- Passwords
- Social security numbers
- Internal reference numbers
Enter field names (comma-separated) in the "Exclude Fields" setting. Field matching is case-insensitive.
Fallback Configuration
If the AI fails to generate a reply (API error, rate limit, etc.), you can configure a fallback static response:
Enable Fallback
Toggle on to send a static message when AI fails
Fallback Subject
Email subject for the fallback message
Fallback Message
Static message to send when AI fails
When AI fails, you'll also receive a notification email with details about the error.
Rate Limits
Configure rate limits to control usage and prevent abuse:
| Plan | Default Per Hour | Default Per Day |
|---|---|---|
| Pro | 100 | 500 |
| Advanced | 500 | 2,000 |
You can set custom limits lower than the defaults in your settings to further control usage.
Viewing AI Reply Logs
Monitor your AI reply activity in the AI Reply Logs dashboard:
- View success/failure status of each AI reply
- Track token usage for each request
- See error details for failed requests
- Filter by success/failure status
- Logs are retained for 14 days
Supported Models
Static Forms AI Reply supports the latest models from all major AI providers. For detailed pricing information, please visit the official documentation for each provider.
OpenAI
gpt-5.2- Latest flagship model for coding and agentic tasks (400K context)gpt-5.2-pro- Maximum intelligence for high-stakes research (400K context)gpt-5-mini- Fast and cost-efficient for well-defined tasksgpt-5- Previous generation flagship model
View OpenAI model details and pricing →
Anthropic
claude-opus-4-5-20251101- Most capable Claude modelclaude-sonnet-4-5-20250929- Best for coding and complex tasks
View Anthropic model details and pricing →
gemini-3-flash- Fast with strong reasoning capabilitiesgemini-3-pro- Complex agentic workflows and reasoninggemini-2.5-pro- Stable with adaptive thinkinggemini-2.5-flash- First stable 2.5 Flash model
View Google Gemini model details and pricing →
Model Selection
Choose models based on your specific needs. Latest generation models (GPT-5, Claude Opus/Sonnet 4.5, Gemini 3) offer improved performance and capabilities. Legacy models remain available for backward compatibility.
Privacy and Security
Data Processing Disclosure
When AI Reply is enabled, form submission data is sent to your configured AI provider (OpenAI, Anthropic, or Google) for processing. This includes form field values (except excluded fields) and your knowledge base content. Please review your AI provider's privacy policy and ensure compliance with your data protection requirements.
API Key Encryption
Your API key is encrypted using AES-256-CBC before storage
Field Exclusion
Sensitive fields can be excluded from AI processing
You Control Your Data
Data is sent directly to your AI provider account, not stored by Static Forms
System Guardrails
To ensure safe, compliant, and professional email responses, Static Forms applies system guardrails to all AI-generated replies. These guardrails are prepended to your custom system prompt. Some guardrails are mandatory and always enforced, while others can be customized to fit your needs.
Mandatory Guardrails
These guardrails are always enforced and cannot be customized or disabled. They ensure proper email formatting and scope limitations:
Email Standards
The AI will never generate:
- Harmful, offensive, discriminatory, or inappropriate content
- Profanity, hate speech, or harassment
- Adult or explicit content
Scope Limitations
The AI will:
- Only answer based on information in your knowledge base
- Acknowledge when information is not available
- Never make up information or hallucinate facts
- Stay focused on the form submission topic
These guardrails work alongside your custom system prompt to ensure all AI-generated emails meet professional and legal standards while maintaining the tone and style you specify.
Optional Guardrails
These guardrails are recommended defaults but can be customized in your settings:
Safety Rules
The AI will never generate:
- Harmful, offensive, discriminatory, or inappropriate content
- Profanity, hate speech, or harassment
- Adult or explicit content
Legal and Compliance Rules
The AI will never:
- Make unauthorized promises, guarantees, or commitments
- Provide medical, financial, or legal advice
- Claim to be a licensed professional
- Share personal information beyond what's in the form submission
Customizing Optional Guardrails
You can customize the optional guardrails (safety and legal/compliance rules) in your AI Reply settings. The guardrails editor allows you to:
- Modify existing rules to match your specific requirements
- Add additional rules or guidelines
- Use the "Use Default" button to restore the default optional guardrails
- Leave blank to use the default optional guardrails
Note
Mandatory guardrails (email formatting and scope limitations) are always enforced and cannot be customized. Only optional guardrails (safety and legal/compliance rules) can be modified.
Important
Guardrails must be plain text format (no MDX, HTML, or markdown) to ensure email compatibility. The AI will generate complete, ready-to-send email content without any placeholders or template variables.
AI Reply vs Auto Responder
Choose the right feature for your needs:
Auto Responder
- Static, templated message
- Same response for all submissions
- Supports template variables
- No external API required
- Instant, no latency
AI Reply
- Dynamic, contextual responses
- Unique reply for each submission
- Uses knowledge base context
- Requires AI provider API key
- Slight latency (~1-3 seconds)
Troubleshooting
AI Reply Not Sending
- Verify AI Reply is enabled in settings
- Check that your API key is valid
- Ensure form includes an email field
- Check AI Reply logs for errors
- Verify you haven't exceeded rate limits
Invalid API Key Error
- Re-enter your API key in settings
- Verify the key is active in your provider's dashboard
- Check that you have billing set up with the provider
- Ensure the key has the required permissions
Poor Quality Responses
- Add more detail to your knowledge base
- Include specific FAQs and answers
- Try a more capable model (e.g., GPT-4o, Claude 3.5 Sonnet)
- Adjust the system prompt with clear instructions
Rate Limit Exceeded
- Wait for the next hour/day for limits to reset
- Increase rate limits in settings (within tier limits)
- Consider upgrading to Advanced plan for higher limits