Snaphost MCP
Deploys projects straight from an AI agent to Snaphost and returns a public URL, with automatic Dockerfile generation, build/run/log management, and deploy lifecycle control.
README
Snaphost MCP server
Deploy the project you are working on straight from your AI agent (Claude Code, Cursor, any MCP client) and get a public URL back.
No Dockerfile required: if the project does not ship one, Snaphost generates it. The folder is packed locally, uploaded, built, and run — the agent gets back a URL it can hand you.
Install
npm install -g @snaphost/mcp
Or skip the install and let your agent run it through npx (see below).
Setup
-
Mint an API key in the Snaphost dashboard under API keys, or
POST /api/v1/keys. It is shown once and looks likesk_.... -
Register the server with your agent.
Claude Code:
claude mcp add snaphost \ --env SNAPHOST_API_KEY=sk_... \ -- npx -y @snaphost/mcpCursor (
.cursor/mcp.json) — or any client using the same format:{ "mcpServers": { "snaphost": { "command": "npx", "args": ["-y", "@snaphost/mcp"], "env": { "SNAPHOST_API_KEY": "sk_..." } } } } -
Tell the agent to deploy: "deploy this project to Snaphost".
Environment
| Variable | Required | Default |
|---|---|---|
SNAPHOST_API_KEY |
yes | — |
SNAPHOST_API_URL |
no | https://api.snaphost.ru |
Tools
| Tool | What it does |
|---|---|
snaphost_deploy |
Pack the folder, upload, build, wait, return the public URL. |
snaphost_status |
Deploy status; URL when running. |
snaphost_logs |
Build/runtime logs (paged via next_since). |
snaphost_list |
Recent deploys. |
snaphost_delete |
Stop and delete a deploy. |
What gets uploaded
snaphost_deploy packs the folder into a tar.gz in memory, honouring
.gitignore and .snaphostignore. These are dropped unconditionally,
whatever the ignore files say:
.git node_modules .env .env.* *.local.env .DS_Store
dist build .next .nuxt coverage __pycache__ *.pyc .venv venv
Symlinks are skipped — the server rejects them. The upload is capped at 50 MB compressed by default; the ignore rules keep a typical project far below that.
Add a .snaphostignore to exclude more. .env files never leave your
machine, so anything the deployed app needs at runtime has to be configured
on the platform side.
If the project ships its own Dockerfile
The runtime injects a PORT environment variable and invokes the container
on it. A server hard-bound to a fixed port builds and deploys fine but never
receives traffic. Make it listen on process.env.PORT (or the equivalent),
falling back to a local default.
Development
npm install
npm run build # tsc -> dist/
npm run typecheck
Requires Node 20+.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.