Claude Web Fetch

Fetch and process URL content using Claude.

Overview

Claude Web Fetch retrieves content from a URL, then sends the fetched content to Anthropic’s Claude Messages API (POST /v1/messages) for analysis or summarization. The URL content is provided as context in the user message alongside a customizable instruction (default: “Summarize this content”). Supports configurable model selection, system prompt, temperature, max tokens, and a content truncation limit of 100,000 characters. Returns the Claude response text and the source URL in configurable output fields.

Category: AI
Tool Name: claude_web_fetch
Version: 1

Appearance: Icon: anthropic | Color: #d4a574

Node Type

Action — processes input items and produces output

Input / Output

Direction	Port(s)
Input	Input
Output	Output, Error

Credentials

This tool requires Anthropic credentials. See the Credentials Guide for setup instructions.

Parameters

Parameter	Type	Required	Default	Description
Model	options	No	Claude Sonnet	The Claude model to use for processing the URL content. Always uses the latest version (auto-updated).
				Options: Claude Opus (most capable — complex analysis and long-form content), Claude Sonnet (balanced — strong quality at lower cost and latency), Claude Haiku (fastest — quick summaries and high-volume processing). Each option tracks the current release of its tier, so the underlying model ID updates without any change to your node.
URL	string	Yes	—	The URL to fetch content from. Falls back to item.json.url if empty. Supports expressions like {{ $json.url }}.
User Message	string	No	Summarize this content	Instruction for what Claude should do with the fetched content. Default: “Summarize this content”. Supports expressions.
System Prompt	string	No	—	Optional system prompt to set Claude’s behavior when processing the URL content. Supports expressions.
Options	collection	No	{}	Advanced generation and output settings.
— Max Tokens	number	No	4096	Maximum number of tokens to generate in the response.
— Temperature	number	No	1	Sampling temperature (0-1). Lower values produce more focused, deterministic output.
— Response Field Name	string	No	response	The output field name where the Claude response text will be stored.
Include Input	boolean	No	false	Whether to include the original input item fields in the output alongside the response.
Max Concurrency	number	No	5	Maximum number of items to process concurrently. Lower than typical due to URL fetching overhead.

Output Data

One fetch and one API call per input item, and one output item per input item. Claude’s analysis lands on the field named by Response Field Name (response by default). Binary data on the input item is forwarded unchanged, and with Include Input on the original item fields are merged in alongside the result.

{
  "response": "Claude's analysis of the fetched page",
  "sourceUrl": "https://example.com/article",
  "contentLength": 48213,
  "wasTruncated": false,
  "model": "claude-...",
  "usage": { "input_tokens": 12480, "output_tokens": 604 },
  "stopReason": "end_turn"
}

• sourceUrl is the URL that was actually fetched, after the item fallback is applied — useful when the URL comes from upstream data.
• contentLength is the character count of the page content that was retrieved.
• wasTruncated is true when the page exceeded the 100,000-character limit and only the leading portion was sent to the model.
• model is the model that actually answered, as reported by Anthropic.
• usage is the token accounting Anthropic returned for the call.
• stopReason says why generation stopped — for example end_turn (finished naturally) or max_tokens (hit the Max Tokens cap).

Reference the result downstream by expression, e.g. {{ $json.response }}.

Usage Examples

• Summarize a web page using Claude
• Extract key information from a URL
• Analyze the content of a documentation page
• Fetch and process blog posts for content review
• Convert web page content into structured data

Example Configuration

Minimal — summarize one page:

{
  "type": "claude_web_fetch",
  "parameters": {
    "url": "https://example.com/article",
    "userMessage": "Summarize this content"
  }
}

Structured extraction from a research page:

{
  "type": "claude_web_fetch",
  "parameters": {
    "url": "{{ $json.url }}",
    "userMessage": "Extract the abstract, key findings, and methodology from this research paper.",
    "systemPrompt": "You are a scientific research assistant. Return structured, factual summaries only.",
    "maxConcurrency": 3,
    "options": {
      "maxTokens": 8192,
      "temperature": 0.1,
      "responseFieldName": "paperSummary"
    }
  }
}

High-volume page classification, with the URL supplied by upstream items:

{
  "type": "claude_web_fetch",
  "parameters": {
    "url": "",
    "userMessage": "Classify this page as one of: blog, news, product, documentation, or other. Respond with only the category label.",
    "systemPrompt": "You are a webpage classifier. Output only the category label, nothing else.",
    "includeInput": true,
    "maxConcurrency": 5,
    "options": {
      "maxTokens": 64,
      "temperature": 0,
      "responseFieldName": "pageCategory"
    }
  }
}

Leaving URL empty makes the node read url from each incoming item, which is the usual pattern when a previous step produced the list of links.

Error Handling

Mode	Behavior
stop	Halts workflow on first error
continue	Skips failed items, passes successful ones through
errorPort	Routes failed items to Error output port

Tips

Claude Web Fetch retrieves content from any URL using Node.js fetch and submits it to the Claude Messages API for analysis or summarization via a customizable instruction. Use it when a workflow needs to automatically extract insights, summaries, or structured data from web pages and remote documents in a single step. Outputs a response text field containing the Claude analysis and a source URL field confirming the content origin, with configurable model, system prompt, temperature, and max tokens.