# webscraperapi full context

## Product summary

webscraperapi is a toolkit for AI agents and developers.

Two access modes:
1) MCP server (one-command option for compatible clients)
2) REST API endpoints (API key based HTTP access)

## MCP setup

### Claude Code

```bash
claude mcp add --transport http webscraperapi https://api.webscraperapi.ai/mcp
```

After adding, make a tool request such as:
- "Use webscraperapi to fetch https://example.com and summarize it."

On first MCP tool call, OAuth opens and user authorization is required.

### Generic MCP config

```json
{
  "mcpServers": {
    "webscraperapi": {
      "type": "http",
      "url": "https://api.webscraperapi.ai/mcp"
    }
  }
}
```

## REST API auth

Pass your key in `Authorization: Bearer YOUR_API_KEY` for direct HTTP calls.

Example:

```bash
curl -G "https://api.webscraperapi.ai/v2/scrape" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d "url=https://news.ycombinator.com" \
  -d "output=markdown"
```

## Available tools/endpoints

- `/v2/scrape`
- `/v2/google_search`
- `/v2/google_shopping_search`
- `/v2/google_ads`
- `/v2/detect_technologies`
- `/v2/serp_domain_frequencies`
- `/v2/verify_email_address`

## Canonical references

- Docs: https://www.webscraperapi.ai/docs
- MCP quickstart: https://www.webscraperapi.ai/docs/mcp
- API reference: https://www.webscraperapi.ai/docs/api-reference
- OpenAPI: https://www.webscraperapi.ai/openapi.json