Skip to main content
The Text Generator agent enables you to generate text using state-of-the-art language models from OpenAI (GPT) and Anthropic (Claude). This agent is perfect for content creation, text transformation, summarization, and any task requiring natural language generation.

​
Use Cases

  • Generate email responses based on incoming messages
  • Summarize long documents or commits
  • Create content from templates with dynamic data
  • Transform data into human-readable formats
  • Perform text analysis and classification

​
Configuration

​
API Authentication

​
apiKey
string
required
Your OpenAI or Anthropic API key, depending on the selected model.

​
Model Selection

​
model
string
required
The AI model to use for text generation. Available options:OpenAI Models:
  • gpt-3.5-turbo - Fast and cost-effective
  • gpt-4 - Most capable GPT-4 model
  • gpt-4-turbo - Faster GPT-4 variant
Anthropic Models:
  • claude-3-haiku-20240307 - Fastest, most compact model
  • claude-3-sonnet-20240229 - Balanced performance and speed
  • claude-3-opus-20240229 - Most powerful Claude model

​
Prompt Configuration

​
prompt
string
required
The text prompt that instructs the AI model on what to generate. You can include dynamic fields from previous workflow steps by clicking on available input fields.Example:
Summarize the following email:

Subject: {{emailSubject}}
From: {{emailFrom}}
Body: {{emailBody}}

Provide a concise 2-3 sentence summary.

​
Input Requirements

The Text Generator can access data from any previous agent in the workflow. Click on the input structure display to see available fields and insert them into your prompt.

​
Output Structure

​
text
string
required
The generated text content from the AI model.

​
Example Output

{
  "text": "This is the generated text from the AI model based on your prompt."
}

​
Usage Example

​
Email Auto-Responder

Configure a workflow that reads emails and generates personalized responses:
  1. Gmail Reader Agent - Reads incoming emails
  2. Text Generator Agent - Configured with: 
    Model: gpt-4-turbo
Prompt: "Write a professional response to this email:

From: {{emailFrom}}
Subject: {{emailSubject}}
Body: {{emailBody}}

Be helpful, concise, and maintain a friendly tone."
  3. Gmail Sender Agent - Sends the generated response

​
Commit Summarizer

Generate summaries of GitHub commits: 
Model: claude-3-sonnet-20240229
Prompt: "Analyze these recent commits and provide a brief project update:

{{commits}}

Format the summary in 3-4 bullet points highlighting key changes."

​
Best Practices

​
Prompt Engineering

  • Be specific and clear in your instructions
  • Provide examples of desired output format
  • Use structure (bullet points, numbered lists) in your prompts
  • Test prompts with the “Test Agent” button before deploying

​
Model Selection

  • Use GPT-3.5 Turbo for simple tasks and faster responses
  • Use GPT-4 or Claude Opus for complex reasoning and analysis
  • Use Claude Haiku for speed-critical applications

​
Cost Optimization

  • Keep prompts concise to minimize token usage
  • Use the most appropriate model for your task (don’t use GPT-4 for simple tasks)
  • Test thoroughly to avoid unnecessary API calls

​
Error Handling

Common errors and solutions:
  • Invalid API Key: Verify your API key is correct and has not expired
  • Rate Limit Exceeded: Wait a moment and try again, or upgrade your API plan
  • Prompt Too Long: Reduce the amount of context in your prompt
  • Model Not Available: Ensure your API key has access to the selected model

​
Testing

​
Test Agent

Tests the text generation with your current prompt configuration. This runs the agent in isolation without executing the entire workflow.

​
Test Workflow

Executes the complete workflow from the beginning, allowing you to see how real data flows through the text generator.

Build docs developers (and LLMs) love

Get started for freeTalk to us