Free Tool

URL to Markdown

Fetch any public web page and get back clean, readable Markdown. Built for AI agents, developer scripts, and anyone who needs web content without the HTML noise.

Markdown
JSON

REST API

Free, CORS-open, no API key required. Use it from your scripts, agents, or directly in an MCP config.

AI Agents

Give your Claude or GPT agent the ability to read any URL as clean context.

RAG Pipelines

Fetch and chunk web pages for vector stores without a browser dependency.

Dev Scripts

Quickly pull article text, docs, or changelogs into your pipeline.

MCP Tools

Call from any MCP-compatible agent to read live web content on demand.

GET  /api/fetch?url=<encoded-url>

curl "https://hatchloop.dev/api/fetch?url=https%3A%2F%2Fexample.com"

POST  /api/fetch

curl -X POST https://hatchloop.dev/api/fetch \ -H "Content-Type: application/json" \ -d '{"url":"https://example.com"}'

Response fields

FieldTypeDescription
urlstringOriginal URL requested
finalUrlstring?URL after redirects (omitted if same as url)
httpStatusnumberHTTP status code from origin server
contentTypestringContent-Type header from origin
markdownstringCleaned Markdown content (max 50,000 chars)
wordCountnumberApproximate word count of extracted text
truncatedbooleanTrue if content was trimmed at 50k chars
fetchedAtstringISO 8601 timestamp of the fetch

Use in an MCP config

{ // Add to your claude_desktop_config.json mcpServers block "hatchloop-tools": { "command": "npx", "args": ["-y", "mcp-remote", "https://hatchloop.dev/mcp"] } }

The MCP endpoint at /mcp exposes fetch_as_markdown as a native tool your agent can call directly — no curl required.