# web-data MCP server

> Provides quality-gated web scraping via Apify with schema validation, quality scoring, and RAG-ready document chunking for AI agents.

- Repository: https://github.com/fctpe/web-data-mcp
- Page: https://hostedmcp.org/servers/fctpe/web-data-mcp
- Transport: —
- Stars: 1
- License: MIT

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `APIFY_TOKEN` | no | no |  |
| `OPENAI_API_KEY` | no | no |  |
| `OTEL_EXPORTER_OTLP_ENDPOINT` | no | no |  |
| `WEB_DATA_MCP_HTTP_TOKEN` | no | no |  |

## Tools (7)

- `scrape_url`
- `run_actor`
- `get_run_status`
- `fetch_dataset_items`
- `validate_dataset`
- `retry_low_quality_run`
- `dataset_to_rag_documents`

## FAQ

### How does the server handle bad scrapes or bot detection?

The server automatically retries with escalating anti-blocking settings, such as residential proxies and browser crawlers, when quality scores are low.

### Can I use this for RAG applications?

Yes, it includes a tool to emit embedding-ready chunks with source attribution and stable content-hash IDs.

### Does it support custom data validation?

Yes, you can validate datasets against your own JSON Schema to check for pass rates and completeness.
