ai-factory-server

Expose a local HTTP API for browsing Claude.ai skills and project files.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/alexanderjamesmcleod/makecents --skill ai-factory-server
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-factory-server
Source: https://github.com/alexanderjamesmcleod/makecents/tree/main/src/skills/business/ai-factory-server.skill
Command: npx skills add https://github.com/alexanderjamesmcleod/makecents --skill ai-factory-server

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires express, ws, and includes scripts (resource) components.

What problem does it solve?

The Skill provides a local, secure API gateway to explore Claude.ai skills and access related project files without exposing a remote service.

Core Features & Use Cases

  • API access to browse and read skills
  • Read-only command execution and context storage in a controlled environment
  • Use Case: Local development workflows that require Claude.ai to inspect skill catalogs and files quickly.

Quick Start

Start the tunnel with ai-factory-tunnel start and call the API at http://localhost:3456.

Frequently Asked Questions about ai-factory-server

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

FAQPage Schema
How do I set up a local API server for Claude.ai to access project files?

A local Express server provides a secure HTTP API for Claude.ai to browse skills and access project files. It features token-based authentication and restricts file path access to ensure your private workspace remains isolated during local development and testing.

Does Claude.ai need token-based authentication to access a local Express server?

Yes, token-based authentication secures the local Express API server, preventing unauthorized access when exposing your project files. This mechanism ensures that Claude.ai interactions with your private workspace remain protected during local development workflows.

How do I restrict file path access when exposing a local API for Claude.ai?

Path restrictions are enforced by the local Express server to confine Claude.ai's read-only file access to approved directories. A designated writable .ai-factory area is also provided for controlled operations, ensuring the broader private workspace remains protected.

What is the best way to browse Claude.ai skills locally during development?

Using a local Express API server is the best way to browse Claude.ai skills locally, providing a secure gateway to inspect skill catalogs and read project files. This approach avoids exposing a remote service while satisfying secure tunneling requirements for private workspaces.

Can I execute commands through a local HTTP API when testing Claude.ai skills?

Yes, the local Express API supports read-only command execution and context storage within a controlled environment. This allows Claude.ai to perform safe, limited operations while testing skill interactions against your local project files during development workflows.