helmdeck

Orchestrate browser, web, data, and code actions via MCP tools.

4|5|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/tosin2013/helmdeck --skill helmdeck
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: helmdeck
Source: https://github.com/tosin2013/helmdeck/tree/main/skills/helmdeck
Command: npx skills add https://github.com/tosin2013/helmdeck --skill helmdeck

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

Helmdeck provides a self-hosted, containerized platform that exposes 36 capability packs as MCP tools, enabling teams to run browser automation, web scraping, filesystem and git operations, document processing, and code execution entirely offline or within a private network.

Core Features & Use Cases

  • 36 capability packs including browser, web scraping, slides, repo orientation, filesystem, git, GitHub, HTTP, vision, OCR/parse, Python/Node execution
  • deterministic, MCP-based tool contracts that can be composed into end-to-end workflows
  • supports optional resources (scripts/, references/, assets/) loaded on demand to extend capabilities

Quick Start

Install Helmdeck locally using the provided install script, then connect to http://localhost:3000 and begin invoking the helmdeck__ tools via MCP.

Frequently Asked Questions about helmdeck

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run browser automation and web scraping locally with an AI agent?

You can run browser automation and web scraping locally by deploying a self-hosted, containerized platform that exposes these actions as MCP tools. Helmdeck provides 36 capability packs for deterministic, offline AI agent workflows without external dependencies.

What is an MCP-compatible catalog for local AI agent workflows?

An MCP-compatible catalog is a unified interface that exposes deterministic tool contracts for AI agents. Helmdeck orchestrates browser, web, data, and code actions through 36 capability packs invoked by helmdeck__ tool names for self-hosted deployments.

Can I execute Python and Node code offline within a private network using MCP tools?

Yes, you can execute Python and Node code entirely offline within a private network. Helmdeck operates as a self-hosted, containerized platform exposing code execution packs as MCP tools for local AI agent workflows.

Does Helmdeck support filesystem and git operations for automated repo orientation?

Yes, Helmdeck supports filesystem, git, and GitHub operations alongside repo orientation. These capabilities are included in the 36 MCP-compatible packs designed for local AI agent orchestration and deterministic workflow composition.

How do I extend AI agent capabilities with custom scripts and assets on demand?

You extend AI agent capabilities by loading optional resources from scripts, references, and assets directories on demand. Helmdeck supports these directories to expand the 36 default capability packs for self-hosted deployments.

What are the limitations of using a self-hosted MCP tool catalog for web scraping?

The primary constraint of a self-hosted MCP tool catalog is that it requires local infrastructure and containerized deployment. Helmdeck operates entirely offline or within a private network, meaning external cloud scaling is unavailable.