---
title: Webfold documentation
description: Read how to install Webfold when it opens, scrape and crawl the web into clean Markdown, run the MCP and CDP servers, and call the hosted API.
---

Webfold turns messy websites and documents into clean text that people and AI can use. It runs as a small program on your own machine, with a hosted API in early access. The program has no public download yet.

**[Quickstart](/docs/getting-started/quickstart)**

Scrape your first page in a few commands.

**[Guides](/docs/guides/scraping)**

Scraping, rendered pages, crawling, search, documents, WARC, CDP and MCP.

**[API reference](/docs/api-reference)**

Every endpoint with parameters, schemas, code samples and a Try it panel.

**[CLI reference](/docs/reference/cli)**

Every subcommand and its arguments.

**[OCR and documents](/docs/guides/ocr)**

Office files, ebooks and PDFs to Markdown, with OCR for scans.

**[PII and secrets](/docs/guides/pii)**

Detect and redact personal data and credentials, with an HTTP service.

**[Corpusforge and Glossa](/docs/guides/corpusforge)**

Governed corpora, and localisation that lives in your repository.

**[Computer use](/docs/guides/computer-use)**

Sandboxed desktops an agent drives and a person can take over.

## Try the API

Open the [playground](https://webfold.ai/playground) to run scrape, crawl, map, search and the PII service and see the results rendered, or build the commands for the local tools. The [API explorer](/docs/api-explorer) is for raw requests, and the [PII API](/docs/pii-api) documents the redaction service.