uercle
Sign In

Beta: This service is in beta and may not function 100% as expected. For issues or feedback, DM @liran_yo on X

Web Fetch & Search API
for AI Agents

Stop wasting tokens on junk. Get exactly what your LLM needs.

Handles JavaScript sitesPrompt-guided extractionCitations included
Start with just $5 — no subscription required

Fetch: Clean Content Extraction

fetch(url="https://example.com/quantum-news", prompt="What are the key points?")
❌

Other Tools Waste Your Money

"Smart" scrapers dump junk on you!

# Site Navigation
- Home
- About
- Products

[Advertisement]

[Cookie Banner]
## Article Title   ← here!
The actual content...

## Related Articles
- Link 1
- Link 2

[Advertisement]

## Footer
© 2024 Company

JavaScript pages? You get nothing!

<html>
<body>
  <div id="root"></div>
  <script src="/bundle.js">
  </script>
</body>
</html>

<!-- That's it.
     Content is rendered
     by JavaScript.
     You get an empty div. -->

Raw HTML wastes tokens!

<html>
 <head>
  <meta charset="utf-8"/>
  <link href="/styles.css"/>
  <script src="/ads.js"/>
  <script src="/track.js"/>
 </head>
 <body>
  <nav><!-- nav --></nav>
  <aside><!-- ads --></aside>
  <main>Content</main> ←
  <footer>...</footer>
 </body>
</html>
✅

Quercle: Clean Extraction

Clean content automatically tailored to your prompt

The main breakthrough in quantum computing came from a new error
correction technique that reduces noise by 40%.
Researchers at MIT developed this approach
using [topological qubits](https://example.com/quantum-breakthrough).

Key points:
• Error rates dropped from 2.3% to 1.4%
• Scales to 100+ qubit systems
• Compatible with existing hardware

This enables practical quantum algorithms for drug discovery and
cryptography within 2-3 years.
See [quantum timeline](https://example.com/quantum-timeline)

Search: Synthesized Answers

search(query="AI breakthroughs 2024")  # optional: allowed_domains, blocked_domains
❌

Other Search Tools Give You Homework

Here's links - go fetch them yourself!

Result 1:
Title: "AI Developments 2024"
Link: https://example.com/article1

Result 2:
Title: "Latest AI Breakthroughs"
Link: https://example.com/article2

Result 3:
Title: "AI Research Updates"
Link: https://example.com/article3

Result 4:
Title: "Machine Learning Progress"
Link: https://example.com/article4

...now go fetch & read them all

AI synthesis with zero specifics!

AI has made significant progress
in 2024.

There have been developments in
language models and other areas.





← zero actual information
✅

Quercle: Synthesized Answer

Synthesized answer with full citations

Answer:
Major breakthroughs in AI developments at 2024 include:

1. Large Language Models: New architectures achieved 40% better
   reasoning through constitutional AI techniques [1]

2. Multimodal AI: Vision-language models process video in real-time,
   enabling autonomous vehicles to medical diagnosis [2]

3. Training Efficiency: Novel optimization reduced training costs
   by 60% while maintaining performance [3]

Sources:
[1] LLM Reasoning Breakthrough
      https://example.com/llm-reasoning-breakthrough
[2] Multimodal Video Processing
      https://example.com/multimodal-video-processing
[3] Efficient Training Methods
      https://example.com/efficient-training-methods

Quick Integration

Use the API directly in your code, wrap it as a tool for your custom agent, or connect via MCP for instant integration with AI tools like Claude Code.

Fetch

Clean, structured content ready for your LLM context.

import requests

response = requests.post(
  "https://quercle.dev/api/v1/fetch",
  headers={"X-API-Key": "YOUR_API_KEY"},
  json={
    "url": "https://example.com",
    "prompt": "Summarize the main points"
  }
)
print(response.json()["result"])

Search

Synthesized, citation-backed answers with complete page analysis.

import requests

response = requests.post(
  "https://quercle.dev/api/v1/search",
  headers={"X-API-Key": "YOUR_API_KEY"},
  json={
    "query": "latest AI developments",
    "allowed_domains": ["example.com"],  # Optional
    # "blocked_domains": ["foo.com"]  # Optional
    # allowed_domains and blocked_domains can not be used together
  }
)
print(response.json()["result"])

MCP Support — Supercharge Your AI Tools

Give your AI tools real-time web intelligence using the Model Context Protocol (MCP). Instantly empower Claude Code and other MCP-enabled tools with structured web fetches, powerful search, and high-signal output optimized for LLMs.

mcp-config.json
{
  "mcpServers": {
    "quercle": {
      "command": "npx",
      "args": ["-y", "quercle-mcp"],
      "env": {
        "QUERCLE_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Claude Code Quick Setup

Add Quercle to Claude Code with a single command:

terminal
claude mcp add --transport stdio quercle --env QUERCLE_API_KEY=YOUR_API_KEY -- npx -y quercle-mcp

Simple, Transparent Pricing

1 credit per fetch • 4 credits per search • Rate limits apply

Starter

$5*
500 credits
$10/1k
  • ✓500 API credits
  • ✓500 fetches / 125 searches
  • ✓Real-time results
  • ✓Email support

Beta Access

DM @liran_yo on X

Pro

$19*
2,000 credits
$9.5/1k
  • ✓2,000 API credits
  • ✓2k fetches / 500 searches
  • ✓Real-time results
  • ✓Email support

Beta Access

DM @liran_yo on X

Scale

$350*
40,000 credits
$8.75/1k
  • ✓40,000 API credits
  • ✓40k fetches / 10k searches
  • ✓Real-time results
  • ✓Email support

Beta Access

DM @liran_yo on X

Ultimate

$4100*
500,000 credits
$8.2/1k
  • ✓500,000 API credits
  • ✓500k fetches / 125k searches
  • ✓Real-time results
  • ✓Email support

Beta Access

DM @liran_yo on X

* Prices shown exclude sales tax/VAT/GST • Credits valid for 6 months

Need custom volume or enterprise features?

support@quercle.dev