<!-- BusyBot node reference — https://busybot.net/tools/claude-web-search/ -->

> Node: Claude Web Search (`claude_web_search`) · Action · v1
> Category: AI · Credentials: Anthropic (`anthropic`)
> Updated: 2026-08-16

# Claude Web Search

> Search the web using Anthropic Claude models with the web_search tool.

## Overview

Claude Web Search uses the Anthropic Messages API (POST /v1/messages) with the web_search_20250305 tool to perform grounded web searches. Claude searches the web for relevant information and returns a synthesized answer with source citations. The response includes text blocks for the answer and web_search_tool_result blocks containing search result URLs, titles, and snippets. Returns the response text along with an array of source objects.

**Category:** AI  
**Tool Name:** `claude_web_search`  
**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](https://busybot.net/credentials/anthropic/) for setup instructions.

### Parameters

| Parameter | Type | Required | Default | Description |
|-----------|------|----------|---------|-------------|
| Model | `options` | No | `Claude Sonnet` | The Claude model to use for web search. Always uses the latest version (auto-updated). |
| | | | | Options: Claude Opus (most capable — complex research and multi-step reasoning), Claude Sonnet (balanced — strong quality at lower cost and latency), Claude Haiku (fastest — simple lookups and high-volume queries). Each option tracks the current release of its tier, so the underlying model ID updates without any change to your node. |
| System Prompt | `string` | No | — | Optional system prompt to set the model's behavior and context. Leave empty for default behavior. Supports expressions. |
| User Message | `string` | Yes | — | The search query or prompt to send to Claude. Falls back to item.message or item.prompt if empty. Supports expressions like {{ $json.query }}. |
| 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 are more deterministic, higher values more creative. |
| — Response Field Name | `string` | No | `response` | The output field name where the 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 | `10` | Maximum number of items to process concurrently. |

## Output Data

One search per input item, and one output item per input item. The synthesized answer lands on the field named by Response Field Name (`response` by default) and the sources Claude consulted land on `sources`. Binary data on the input item is forwarded unchanged, and with Include Input on the original item fields are merged in alongside the result.

```json
{
  "response": "The synthesized, web-grounded answer",
  "sources": [
    { "url": "https://example.com/article", "title": "Article title", "snippet": "the matched passage" }
  ],
  "model": "claude-...",
  "usage": { "input_tokens": 940, "output_tokens": 512 },
  "stopReason": "end_turn"
}
```

- `sources` is an empty array when the model answered without running a search.
- `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 }}` or `{{ $json.sources }}`.

## Usage Examples

- Search the web for current information using Claude
- Research a topic with source citations via Claude web search
- Find the latest news about a company or technology
- Get up-to-date answers grounded in web sources
- Lookup recent events and return source URLs

## Example Configuration

Minimal — one grounded lookup per item:

```json
{
  "type": "claude_web_search",
  "parameters": {
    "userMessage": "What are the latest advancements in quantum computing?"
  }
}
```

Research with a system prompt and a named output field:

```json
{
  "type": "claude_web_search",
  "parameters": {
    "systemPrompt": "You are a professional research analyst. Provide thorough, cited summaries.",
    "userMessage": "Summarize recent regulatory changes affecting AI companies in the EU.",
    "includeInput": true,
    "maxConcurrency": 5,
    "options": {
      "maxTokens": 8192,
      "temperature": 0.3,
      "responseFieldName": "researchSummary"
    }
  }
}
```

Batch fact-checking a claim carried on each item:

```json
{
  "type": "claude_web_search",
  "parameters": {
    "systemPrompt": "Verify whether the following claim is accurate based on current web sources. Reply TRUE or FALSE followed by a short explanation.",
    "userMessage": "{{ $json.claim }}",
    "includeInput": true,
    "maxConcurrency": 10,
    "options": {
      "maxTokens": 256,
      "temperature": 0,
      "responseFieldName": "verificationResult"
    }
  }
}
```

### 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 Search queries the live web using the Anthropic Messages API with the web_search_20250305 tool to retrieve grounded, up-to-date information. Use this tool when a workflow requires current information beyond a model training cutoff, such as breaking news, live pricing, or recent events. It produces a synthesized response text and an array of source objects, each containing a URL, title, and snippet from the retrieved search results.