DropZoom documentation

DropZoom publishes static websites and everyday files to a link. Drop a folder, a ZIP file or a few files, and you get an HTTPS address anyone with the link can open. People use it from a browser. Agents use it from a small command-line tool or the HTTP API.

DropZoom hosts finished files. It does not build apps, run server code or provide a database.

Two ways in

In a browser. Open publish.dropzoom.link, add your files and publish. Without an account you get a guest link that lasts 24 hours. Register to keep the same address. Start with the quickstart.

As an agent. A person registers, saves the recovery / agent key and gives it to the agent through a secret store. The agent installs the CLI, connects once and publishes. Every publish prints a JSON receipt with the link. Start with the agent quickstart.

What you can publish

• Static websites and built web apps (a root index.html opens as the site)
• Project demos and HTML reports
• PDFs, images, video and other everyday files (a folder without index.html gets a simple file listing)

Links are public unless you set a visitor password. A random address is not private, so only publish content you are happy for anyone with the link to see.

Reference

• Quickstart: publish from a browser, guest links, keeping a link.
• Agent quickstart: install, connect, publish, read the receipt.
• CLI reference: every command and option.
• HTTP API: routes, requests and responses.
• Errors: the error format and every public error code.
• Limits: visibility, expiry, sizes and rate limits.
• Changelog: what changed in each release.

For machines

• Every docs page is also plain Markdown: add .md to the page name, for example /docs/cli.md, or request the page with Accept: text/markdown.
• All pages in one file: llms-full.txt.
• OpenAPI 3.1 description of the HTTP API: openapi.json.
• API catalog: /.well-known/api-catalog.
• How an agent gets access: auth.md.