create-agent-adapter

Bridge an AI runtime to the 1% AI Labs orchestration layer.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/kysonttan-sys/1percent-AI-Labs --skill create-agent-adapter-kysonttan-sys
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-agent-adapter
Source: https://github.com/kysonttan-sys/1percent-AI-Labs/tree/main/.agents/skills/create-agent-adapter
Command: npx skills add https://github.com/kysonttan-sys/1percent-AI-Labs --skill create-agent-adapter-kysonttan-sys

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Technical guide for creating a new 1% AI Labs agent adapter. It provides the blueprint to bridge a chosen agent runtime (e.g., Claude Code, Codex, or a custom process) to the 1% AI Labs orchestration layer, enabling a consistent end-to-end adapter implementation across server, UI, and CLI.

Core Features & Use Cases

  • Defines a self-contained adapter package with server, UI, and CLI modules that integrate with the 1% AI Labs ecosystem.
  • Details the required exports, module structure, and registration patterns so adapters are discoverable by all three consumers.
  • Covers conventions, environment handling, session management, and testing hooks to ensure reliable operation in production workflows.

Quick Start

Create a new adapter package under packages/adapters, implement server/ui/cli modules, and register the adapter in the three registries.

Frequently Asked Questions about create-agent-adapter

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

FAQPage Schema
How do I integrate a new agent runtime into an orchestration layer?

To integrate an agent runtime, build a self-contained adapter package with server, UI, and CLI modules, then register it in the three central registries. This bridges the chosen runtime to the orchestration layer, ensuring discoverability across all consumers.

What components are required to build an agent adapter end-to-end?

An agent adapter requires a root index, server execute and parse modules, UI connectors, and CLI handlers. These components form a complete package that wires the agent runtime into the broader ecosystem for consistent operation.

How do I register an agent adapter so it is discoverable by the server, UI, and CLI?

Register an agent adapter by adding it to the three distinct registries for server, UI, and CLI consumers. This consistent registration pattern ensures the adapter is recognized and callable across the entire orchestration layer.

Can I use a custom process instead of Claude Code or Codex for my agent adapter?

Yes, the agent adapter architecture supports adding a custom process alongside existing runtimes like Claude Code or Codex. The adapter package structure treats the underlying runtime as pluggable, bridging any chosen process to the orchestration layer.

What conventions should I follow for environment handling and session management in an agent adapter?

Agent adapters must follow established conventions for environment handling, session management, and testing hooks. These practices ensure the adapter operates reliably within production workflows and integrates cleanly with the orchestration layer.

Where should I place a new agent adapter package in my project structure?

Create a new agent adapter package under the packages/adapters directory. Implement the required server, UI, and CLI modules within this self-contained package, then register it to complete the integration.