activities

latest

false

Integration Service activities

Last updated May 8, 2026

Generate Message

Description

Given a prompt, generates a message response using the Anthropic Claude Messages API (/v1/messages). Supports plain text, multi-turn conversations, and multimodal inputs.

Project compatibility

Windows | Cross-platform

Configuration

Connection - The connection established in Integration Service. Access the dropdown menu to choose, add, or manage connections.
Model - The Claude model to use for generating the response.
Prompt - The prompt for Claude to use when generating the response.
Max Tokens - The maximum number of tokens allowed for the generated response. Maximum value is 100,000. Default value is 256.

Advanced

Options

System prompt - A system prompt provides context and instructions to Claude, such as specifying a particular goal or role. This is an optional field.
Top P - A number between 0 and 1. The lower the number, the fewer tokens are considered. Defaults to 0.7.
Top K - Used to remove "long tail low probability responses". Only sample from the top K options for each subsequent token.
Temperature - The amount of randomness injected into the response. Accepted values are between 0 and 1.

Output

Response - The full message object returned by the API.
Generated text - The text of the top-choice response generated by Claude.

Frequently asked questions

What prompt formats are supported?

In addition to plain text, the Generate Message activity supports structured conversational and multimodal inputs.

Multiple conversational turns

Pass an array of messages to simulate a multi-turn conversation:

[
  {"role": "user", "content": "Hello, Claude"},
  {"role": "assistant", "content": "Hello!"},
  {"role": "user", "content": "Can you describe LLMs to me?"}
]
[
  {"role": "user", "content": "Hello, Claude"},
  {"role": "assistant", "content": "Hello!"},
  {"role": "user", "content": "Can you describe LLMs to me?"}
]

Putting words in Claude's mouth

Pre-fill the assistant's reply to guide the response format:

[
  {
    "role": "user",
    "content": "What is latin for Ant? (A) Apoidea, (B) Rhopalocera, (C) Formicidae"
  },
  {
    "role": "assistant",
    "content": "The answer is ("
  }
]
[
  {
    "role": "user",
    "content": "What is latin for Ant? (A) Apoidea, (B) Rhopalocera, (C) Formicidae"
  },
  {
    "role": "assistant",
    "content": "The answer is ("
  }
]

Multimodal input (image + text)

Option 1: Base64-encoded image

"messages": [
  {
    "role": "user",
    "content": [
      {
        "type": "image",
        "source": {
          "type": "base64",
          "media_type": "<IMAGE_MEDIA_TYPE>",
          "data": "<IMAGE_BASE64>"
        }
      },
      {
        "type": "text",
        "text": "What is in the above image?"
      }
    ]
  }
]
"messages": [
  {
    "role": "user",
    "content": [
      {
        "type": "image",
        "source": {
          "type": "base64",
          "media_type": "<IMAGE_MEDIA_TYPE>",
          "data": "<IMAGE_BASE64>"
        }
      },
      {
        "type": "text",
        "text": "What is in the above image?"
      }
    ]
  }
]

Option 2: URL-referenced image

{
  "role": "user",
  "content": [
    {
      "type": "image",
      "source": {
        "type": "url",
        "url": "https://upload.wikimedia.org/wikipedia/commons/a/a7/Camponotus_flavomarginatus_ant.jpg"
      }
    },
    {
      "type": "text",
      "text": "What is in the above image?"
    }
  ]
}
{
  "role": "user",
  "content": [
    {
      "type": "image",
      "source": {
        "type": "url",
        "url": "https://upload.wikimedia.org/wikipedia/commons/a/a7/Camponotus_flavomarginatus_ant.jpg"
      }
    },
    {
      "type": "text",
      "text": "What is in the above image?"
    }
  ]
}

For more details, refer to the Anthropic Messages API documentation.

On this page

Description
Project compatibility
Configuration
Advanced
Options
Output
Frequently asked questions
What prompt formats are supported?

Was this page helpful?

PREVIOUSActivities

NEXTGenerate Text Completion (Deprecated)

Integration Service activities

Description​

Project compatibility​

Configuration​

Advanced​

Options​

Output​

Frequently asked questions​

What prompt formats are supported?​

Multiple conversational turns​

Putting words in Claude's mouth​

Multimodal input (image + text)​