mcpField
H
host0

host0

Remote
host0.aiCloud & Infra0Updated 4d ago

About host0

host0’s MCP server manages apps published to host0, including sharing, versions, analytics and shared data. The accompanying skill lets coding agents publish files and folders to a shareable host0.app URL.

Key features

  1. 01Publish files, folders and apps
  2. 02Manage visibility, sharing and URLs
  3. 03Inspect versions and restore deployments
  4. 04View cookieless analytics
  5. 05Use a shared database without server code

Use cases

  1. 01Publish a static site or PDF to a link
  2. 02Share an app privately or publicly
  3. 03Restore an earlier deployment
  4. 04Build and publish an app with shared data

MCP support

The host0 plugin bundles a skill for publishing and an MCP server for managing live projects, including sharing, versions, analytics and shared data.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.host0.ai/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

host0 skills

skills.sh

host0 hosts whatever your agent creates — HTML pages, sites, apps with a shared database, Markdown, PDFs — at a link you can share. This repo holds the host0 skill: it teaches any coding agent to publish a file or folder to a live https://<slug>.host0.app URL in one step.

Install

npx skills add host0-ai/skills

Works with Claude Code, Cursor, Codex, Gemini CLI, GitHub Copilot, Windsurf, Cline and every other agent the skills CLI supports. Add -g to install it for all your projects. npx skills add https://host0.ai installs the same skill straight from host0.

No Node? One curl command installs the same files into your agent's skills folder. It uses no sudo and never overwrites a copy you changed without --force:

curl -fsSL https://host0.ai/install.sh | bash
As a plugin: the skill and the MCP server in one install

The skill publishes; the MCP server manages what's live (sharing, versions, analytics, shared data). The plugin brings both.

Claude Code:

/plugin install host0 --marketplace host0-ai/skills

Codex:

codex plugin marketplace add host0-ai/skills

Then install host0 from the plugin directory. Cursor reads the same plugin.json and mcp.json: add this repository as a plugin from Cursor's plugin settings.

Use it

Ask your agent:

  • "Deploy this to host0"
  • "Give me a link to this PDF"
  • "Build a team todo list with a shared database and publish it"
  • "Roll back the last deploy"

The first publish starts a device login: the agent shows a link and a short code, you approve it on host0.ai in your own browser, and the agent gets an API key. Nothing personal is typed into the chat.

What you get

  • Private by default. Only you and people you invite by email can open an app. Open it to anyone with the link, or make it public on the web, when you want to.
  • A shared database with no server code. Declare record collections in an h0.json and the app's frontend reads and writes them on its own origin.
  • Versions and one-click restore, cookieless analytics, and a readable URL whenever you rename it.
  • Any file or folder: a static site, a single HTML page, a PDF, images, video, Markdown.

There is a free plan. Docs: host0.ai/docs.

Skills

SkillWhat it does
host0Publish files, folders and apps with shared data to *.host0.app; manage visibility, sharing, versions and URLs over the host0 API.

Source

This repo is a mirror. The skill is developed alongside the host0 platform and synced here on every change, so the copy at host0.ai/skill.md and this one are always the same. Report issues at [email protected] or open an issue here.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.