Loom addresses the challenge of efficiently generating and managing API documentation, particularly JSON Schemas. It targets developers and engineers by providing an AI-powered agent that leverages a TUI chat interface to create and update schemas, integrated with a modern web browser for viewing and a dynamic mock service for testing. This streamlines the API development lifecycle, offering a unified tool for documentation and mock data generation.

How It Works

Loom employs an AI-driven approach, using a TUI chat interface to interact with Large Language Models (like DeepSeek or OpenAI) for generating and refining JSON Schemas. It supports entity modeling for reusable schema components, referenced via x-entity-ref. A React-based Single Page Application (SPA) serves as a web browser for interactive schema rendering and documentation viewing. A built-in mock service dynamically generates realistic mock data based on the defined JSON Schemas, enabling immediate API testing.

Quick Start & Requirements

• Installation: Global installation via npm install -g loom or yarn global add loom.
• Environment: Requires Node.js ≥ 18.0.0.
• Prerequisites: A DeepSeek API Key is mandatory for AI features.
• Configuration: A global configuration file (~/.loom/config.json or %APPDATA%/loom/config.json) guides setup on first run.
• Local Development: Standard git clone, npm install, npm run build:all.
• Docs: Links to official quick-start, docs, or demo pages are not explicitly provided in the README.

Highlighted Details

• AI-driven JSON Schema generation via TUI chat.
• Entity modeling for reusable schema components using x-entity-ref.
• Interactive web browser (React SPA) for schema visualization.
• Dynamic mock API service generating data from schemas.
• TUI commands for controlling Mock and Web Viewer services.
• LLM-driven source code API scanning (/scan command) with incremental caching.
• Automatic version upgrade prompts.

Maintenance & Community

The project is developed by "the Loom team." While specific community channels like Discord or Slack are not detailed, users are encouraged to engage via GitHub Issues or Discussions for support and feedback. No information on notable contributors, sponsorships, or partnerships is provided.

Licensing & Compatibility

The README does not specify the project's license. This omission requires clarification for assessing commercial use or closed-source linking compatibility.

Limitations & Caveats

A DeepSeek (or compatible LLM) API key is a mandatory requirement for core AI functionalities. The project relies on LLM performance for schema generation and code scanning, which can introduce variability. The absence of explicit licensing information is a significant caveat for adoption decisions.