Quercle

AI & Machine Learning Utilities

v1.0.0

Last updated Feb 25, 2026

n8n nodes for Quercle AI-powered web search and fetch API

7 Weekly Downloads

51 Monthly Downloads

View on NPM GitHub Repository

Included Nodes

Quercle

Description

@quercle/n8n

AI-powered web search, fetch, and extraction node for n8n.

Install

1. Go to Settings > Community Nodes in your n8n instance
2. Enter @quercle/n8n-nodes-quercle
3. Agree to the risks and install

Alternatively, install manually:

cd ~/.n8n/nodes
npm install @quercle/n8n-nodes-quercle

Restart n8n after manual installation.

Setup

Credentials

1. Get an API key from quercle.dev (starts with qk_)
2. In n8n, go to Credentials > New Credential
3. Search for Quercle API
4. Paste your API key and save

Environment Variable

For self-hosted n8n, you can set the API key as an environment variable instead:

export QUERCLEAPIKEY=qkyourapikeyhere

If both a credential and the environment variable are configured, the credential takes precedence.

Operations

Search

AI-synthesized web search. Returns a natural-language answer with source citations.

Fetch

Fetch a URL and analyze its content with an AI prompt. Returns a processed result based on your instructions.

| Parameter | Type | Required | Description |
|———–|——|———-|————-|
| URL | string | Yes | The URL to fetch |
| Prompt | string | Yes | Instructions for how to process the page content |

Raw Search

Raw web search results without AI synthesis. Returns search results directly in your chosen format.

Raw Fetch

Fetch raw page content from a URL without AI processing.

Extract

Extract relevant content from a URL based on a query. Returns only the parts of the page that match your query.

Usage

All operations return a result field. Operations with the safeguard option also return an unsafe field indicating whether the content was flagged.

Using Search in a workflow

1. Add a Quercle node
2. Set Operation to Search
3. Enter a query, optionally restrict to specific domains
4. The output result contains the AI-generated answer with citations

Operation: Search
Query: {{ $json.searchTerm }}
Domain Filter: Allowed Domains
Domains: github.com, stackoverflow.com

Using Fetch in a workflow

1. Add a Quercle node
2. Set Operation to Fetch
3. Provide a URL and a prompt describing what to extract or summarize

Operation: Fetch
URL: {{ $json.url }}
Prompt: Extract the product name, price, and availability. Return as JSON.

Using Raw Search for structured data

Choose Raw Search with JSON format when you need machine-readable search results for downstream processing.

Operation: Raw Search
Query: {{ $json.topic }}
Format: JSON

Using Raw Fetch for full page content

Choose Raw Fetch when you need the complete page content (e.g., to feed into another AI node).

Operation: Raw Fetch
URL: {{ $json.url }}
Format: Markdown

Using Extract for targeted content

Choose Extract when you only need specific information from a page.

Operation: Extract
URL: https://example.com/pricing
Query: pricing tiers and feature comparison
Format: JSON

Using as an AI Agent tool

The Quercle node has usableAsTool enabled, so you can attach it directly to an n8n AI Agent node. The agent can then call Search, Fetch, or any other operation as part of its reasoning loop.

Troubleshooting

Enable Continue on Fail in the node settings to prevent a single failed item from stopping the entire workflow. Failed items return { "error": "error message" } instead of throwing.

License

MIT