# Claude MCP Skill

The `scraperapi‑mcp` skill teaches your LLM agent how to make the most of the ScraperAPI MCP tools.  Installing the skill provides best-practice guidance, automatic tool selection, and recipes for common web‑data workflows. Without it, you can still call the MCP tools by using the `scrape` keyword and specifying parameters like **url, country\_code, render, etc.**

The skill helps your agent choose between the various ScraperAPI tools (scrape, google\_search, amazon\_search, etc.) based on the task. For example, it suggests using `google_search` for research or `amazon_product` for product details.

### Installation

**Prerequisites:**

* ScraperAPI account.
* Node.js installed to run `npx` commands.
* A client that can load agent skills (for example Claude Desktop with MCP enabled).

{% tabs %}
{% tab title="GitHub" %}

```bash
# Install directly from GitHub
npx skills add scraperapi/scraperapi-skills
```

{% endtab %}

{% tab title="ClawHub" %}

```bash
# Install from ClawHub
npx clawhub install scraperapi-mcp
```

{% endtab %}
{% endtabs %}

### How it Works

When your agent detects that a task matches the skill’s triggers (scraping, Google/Amazon/Walmart/eBay searches, Redfin queries, crawling, etc.), it loads the skill’s guidance into the system prompt. The skill contains:

* A list of **all available** ScraperAPI tools and their parameters.
* A decision tree for selecting the right tool.
* Best practices for cost, pagination, localization and result formatting.
* Guidance on handling large outputs efficiently.

### Prompt Example

<mark style="background-color:blue;">Real-time News & Market Data</mark>

\- *"I need latest Google news for the current price of Gold."*

**Response**

Claude now factors in the `SKILL.md` file when making a decision. Instead of manually parsing the HTML or relying on raw search results, it sends a request to ScraperAPI's **Google News Structured Data endpoint**. That endpoint is specifically designed to return clean, structured JSON data for Google News results

<figure><img src="/files/nSQhl3gRS3tx39pa8h4v" alt=""><figcaption></figcaption></figure>

`ScraperAPI` handles:

* Bypassing the Bot protection mechanisms.
* Fetching the Google News results.
* Structuring the response into machine-readable data.

<details>

<summary>GoldPriceResponse.md</summary>

```markdown
Here are the latest Google News articles on gold prices from the past hour — as of Friday, February 27, 2026:

🥇 Gold Price Right Now

Gold Futures: Surged past $5,200/oz — up ~$65/oz on the day
Silver: Also rallying, up ~$5.75/oz
Driver: Safe-haven demand fueled by U.S.-Iran geopolitical tensions and hotter-than-expected U.S. PPI inflation data


📰 Breaking Headlines (Last Hour)

Gold surges past $5,200 as US-Iran tensions rattle markets (Economic Times – 21 mins ago)
Geopolitical tensions between the US and Iran pushed gold futures above $5,200, fueling strong safe-haven demand. Experts are revising portfolio allocation strategies.
Comex gold rises $65/oz; silver gains $5.75/oz on safe-haven buying (Live Mint – 50 mins ago)
Precious metals held firm in Friday's session as rising Middle East tensions continued to drive safe-haven demand for both gold and silver.
Wall Street slides amid AI fears, hot inflation data, and Iran tensions (Times of India – 26 mins ago)
US stocks fell sharply as gold benefited from a flight to safety — the Dow Jones crashed 777 points, boosting gold's appeal.
US producer inflation higher than expected on services costs (Economic Times – 17 mins ago)
A surprise jump in U.S. PPI data is adding upward pressure on gold as an inflation hedge.


📊 Quick Summary
MetricValueGold (Comex Futures)$5,200+/ozDaily Gain~+$65/ozKey DriversUS-Iran tensions, hot US PPI data, stock market sell-offSilverAlso up +$5.75/ozMarket MoodRisk-off / Safe-haven buying
Gold is surging today driven by a perfect storm of geopolitical risk, inflation fears, and a broad Wall Street sell-off.
```

</details>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.scraperapi.com/integrations/llm-integrations/claude-mcp-skill.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.