create-agent

Create modular AI agents with the OpenRouter SDK and optional Ink TUI.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/zh-impact/fcollections --skill create-agent-zh-impact
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-agent
Source: https://github.com/zh-impact/fcollections/tree/main/.agents/skills/create-agent
Command: npx skills add https://github.com/zh-impact/fcollections --skill create-agent-zh-impact

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This capability enables developers to bootstrap a modular AI agent quickly, providing a standalone agent core, extensible hooks, and optional Ink TUI for interactive use.

Core Features & Use Cases

  • Standalone Agent Core: Runs independently and supports hooks and lifecycle management.
  • OpenRouter SDK Integration: Unified access to 300+ language models via a single API.
  • Optional Ink TUI: Rich terminal UI or headless operation for scripting environments.
  • Tooling and Extensibility: Add tools, HTTP interfaces, or Discord integrations to tailor behavior.

Quick Start

Instantiate a new agent project with your OpenRouter API key and run the headless loop, then interact via the CLI or embed in a UI.

Frequently Asked Questions about create-agent

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

FAQPage Schema
How do I bootstrap a modular AI agent with OpenRouter?

You can bootstrap a modular AI agent with OpenRouter by instantiating a new project with your API key and running the headless loop. This provides a standalone agent core with lifecycle management and extensible hooks for interactive CLI or UI embedding.

Can I run an AI agent headless for scripting environments?

Yes, you can run an AI agent headless for scripting environments. The agent core operates independently and supports an optional Ink TUI, allowing you to choose between a rich terminal UI or automated headless execution.

Does the OpenRouter SDK support multiple language models for AI agents?

Yes, the OpenRouter SDK supports multiple language models for AI agents. It provides unified API access to over 300 different language models, allowing you to easily switch or test models within your modular agent.

How do I add tools or HTTP interfaces to a TypeScript AI agent?

To add tools or HTTP interfaces to a TypeScript AI agent, utilize the agent's extensible hooks and tooling features. You can tailor behavior by integrating HTTP interfaces, Discord integrations, or custom scripts within the supported directories.

What is the best way to manage AI agent lifecycle and hooks?

The best way to manage AI agent lifecycle and hooks is by deploying a standalone agent core. This approach supports lifecycle management and extensible hooks, ensuring modular control over agent behavior and execution flow.

Do I need an OpenRouter API key to create a TypeScript AI agent?

Yes, you need an OpenRouter API key to create a TypeScript AI agent. The quick start process requires you to instantiate the agent project with your OpenRouter API key to access the 300+ language models via the unified API.