By Max Techera · Open-source (MIT) · Updated June 2026
Blog

What is an MCP server, and why does it matter for sharing?

MCP lets AI agents call tools. Here's the plain-English version — and how an MCP 'publish' tool turns agent output into shareable links.

npxnpx drops-cli report.html --managed

The Model Context Protocol (MCP) is a standard way for AI agents to call external tools. An MCP server exposes tools (like publish_html) that the agent can invoke mid-task.

Why it matters for artifact sharing

With an MCP publish tool, your agent can finish a report and immediately hand you a link — no copy-paste. npx -y drops-mcp gives any MCP client exactly that, landing on your own domain.

Add it in one line

claude mcp add drops -- npx -y drops-mcp

FAQ

Does it work with any AI agent?
Yes. It's a CLI plus a SKILL.md and an MCP server, so Claude Code, Codex, Cursor, OpenCode, Gemini CLI — anything that runs a shell or speaks MCP — can use it.
Is it really open-source and self-hosted?
Yes — MIT licensed, and it runs on your own Vercel Blob + domain. No third party ever holds your content (it's encrypted client-side) or controls your URL.

Share what your AI builds — on your own domain.

Open-source, zero-knowledge, free. Try it in one command:

npx drops-cli report.html --managed