What runtime prerequisites does the autopilot require?
Docker must be installed and accessible. 2GB+ RAM is recommended, and swap is strongly recommended on small Linux hosts.
@stakpak
Basic information
Ship your code, on autopilot. An open source agent that lives on your machines 24/7 and keeps your apps running. 🦀
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"cli": {
"command": "docker",
"args": [
"pull",
"ghcr.io/stakpak/agent:latest"
]
}
}
}Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Stakpak Agent CLI is an open-source, security-hardened autonomous agent that lives on your machines 24/7, keeps your applications running, and only pings when it needs a human. It integrates with your DevOps stack to provide AI-powered infrastructure management without giving an LLM direct access to production secrets. It can be used as a secure MCP proxy or expose its tools through an MCP server.
Install via curl (curl -sSL https://stakpak.dev/install.sh | sh), Homebrew (brew tap stakpak/stakpak && brew install stakpak), or Docker (docker pull ghcr.io/stakpak/agent:latest). Initialize with stakpak init, start the 24/7 autonomous runtime with stakpak autopilot up (or stakpak up). Use MCP server modes via --tool-mode local (file/command operations only) or --tool-mode remote (AI-powered). Configure profiles in ~/.stakpak/config.toml and autopilot wiring in ~/.stakpak/autopilot.toml.
Docker must be installed and accessible. 2GB+ RAM is recommended, and swap is strongly recommended on small Linux hosts.
It uses Dynamic Secret Substitution: the AI can read/write/compare secrets without ever seeing the actual values.
Yes. You can configure Anthropic, OpenAI, Gemini, or custom OpenAI-compatible endpoints (e.g., Ollama) in `~/.stakpak/config.toml`.
Yes, it is licensed under Apache 2.0 and available on GitHub.
Local Mode (`--tool-mode local`) allows file operations and command execution without an API key. Remote Mode (`--tool-mode remote`) enables AI-powered tool use.
Set up and manage the Eddyter AI rich text editor by talking to your AI agent — for Claude, Cursor, Codex and other MCP tools. A Model Context Protocol server that lets AI tools add Eddyter to a React, Next.js, Vite or R


Official Porkbun MCP server, maintained by Porkbun. Check domain availability and pricing, register, renew and transfer domains, and manage DNS records, nameservers, URL forwarding, glue records, DNSSEC and domain contac

Agent-Native Terminal

Automated codebase documentation for GitHub: searchable docs, MCP context for AI agents, doc drift detection, and Friday Cleanup PRs.
Comments