afd

Develops CLI-first apps with validated commands using MCP integration for AI agents.

4|1|Updated Dec 30, 2025
One-click install
npx skills add https://github.com/lushly-dev/afd --skill afd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: afd
Source: https://github.com/lushly-dev/afd/tree/main/.claude/skills/afd
Command: npx skills add https://github.com/lushly-dev/afd --skill afd

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill provides patterns and tools for building software where AI agents are first-class users, simplifying agent integration and command design.

Core Features & Use Cases

  • Command Design: Define commands with Zod schemas for robust input validation and clear agent interaction.
  • CLI Validation: Test commands via the command line before building any UI, ensuring reliability.
  • MCP Integration: Set up and connect to MCP servers for agent communication.
  • Use Case: When developing a new application feature, use AFD to first define it as a command-line interface (CLI) command, validate its behavior, and then build the UI or agent integration on top of that solid foundation.

Quick Start

Use the afd skill to define a new command with a Zod schema for input validation.

Frequently Asked Questions about afd

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

FAQPage Schema
How do I design a CLI command for AI agent integration?

Agent-First Development patterns design CLI commands by defining structured inputs with Zod schemas for robust CLI validation, ensuring predictable outputs and reliable agent interaction before UI construction.

What is Agent-First Development and when do I need it?

Agent-First Development (AFD) is a software engineering approach where AI agents are first-class users. You need AFD when building applications requiring seamless human-agent collaboration and structured command execution.

How do I set up an MCP server for agent communication?

Setting up an MCP server for agent communication involves using AFD patterns to configure server integration, facilitating structured communication channels between CLI commands and AI agents.

Can I validate command behavior before building a UI?

Yes, you can validate command behavior before building a UI by testing commands via the command line using CLI validation, ensuring the underlying logic is reliable before adding user interfaces.

Does Zod schema validation work with command-first architecture?

Zod schema validation works with command-first architecture by defining structured inputs that ensure robust validation and clear interaction patterns for agent-ready applications within the AFD workflow.

What are the limitations of building agent-ready applications with CLI validation?

CLI validation for agent-ready applications requires defining strict Zod schemas upfront, meaning any input format deviations will fail validation, necessitating precise command design before agent execution.