Overview

The Claude node allows you to use Anthropic’s AI models inside your Flowmage flows.

Similar to the OpenAI node, Claude can generate text, analyze data, and assist with complex reasoning tasks—making your automations smarter and more flexible.

When to Use This Node

Use the Claude node when you want to:

• Generate or rewrite text

• Summarize large inputs

• Analyze structured or unstructured data

• Classify or extract information

• Perform advanced reasoning tasks

? Claude is especially strong in:

• Long-form content

• Structured reasoning

• Consistent formatting

Configuration

Model

Select the Claude model you want to use.

Example:

Claude Sonnet 4.6

? Different models may vary in performance, speed, and cost.

Reasoning Effort

Controls how much processing the model applies.

Options:

• Low → Fast, simple tasks

• Medium → Balanced (recommended)

• High → Complex reasoning and analysis

Example:

Medium

API Key (Optional)

Provide your own Anthropic API key if needed.

• Leave empty → uses system default

• Fill in → uses your own account

Example:

sk-ant-...

Instructions

Defines the role and behavior of the AI.

Think of this as:
? “How should the AI behave?”

Example:

You are an assistant that extracts structured data from text and returns clean JSON.

Input

The data you send to the AI.

You can:

• Enter static text

• Use Flowmage variables

• Combine both

Example:

Extract key information from this text:
@webhook.description@

How It Works

1. Flow reaches the Claude node

2. Instructions define behavior

3. Input is sent to the model

4. Claude processes the request

5. A response is generated

6. Output becomes available for next steps

Output / Variables

The Claude node returns generated content.

Example usage:

@sleep.output@

? sleep = your Output ID

Example Use Cases

Summarization

Instructions:
Summarize the following text in a professional tone.

Input:
@document.content@

Data Extraction

Instructions:
Extract name, email, and phone number as JSON.

Input:
@formData@

Content Generation

Instructions:
Write a friendly follow-up email.

Input:
@customer.message@

Classification

Instructions:
Classify this ticket as HIGH, MEDIUM, or LOW priority.

Input:
@ticket.description@

Common Mistakes

❌ Unclear Instructions

Claude may return unexpected results.

✅ Be specific about:

• Output format

• Tone

• Structure

❌ Missing Input Data

The AI lacks context.

✅ Always pass relevant variables

❌ Overly Complex Prompts

Too many instructions can reduce clarity.

✅ Keep prompts simple and structured

❌ Not Validating Output

AI output is not always guaranteed.

✅ Use conditions to validate responses

Pro Tips

• Use Claude for longer or more complex inputs

• Ask for structured output (JSON) when automating

• Combine with conditions to act on results

• Reuse prompts across flows for consistency

• Compare results with OpenAI node for best fit

Summary

The Claude node allows you to integrate powerful AI capabilities into your flows using Anthropic models.

It is ideal for complex reasoning, structured outputs, and handling larger inputs within your automations.