ワンクリックでManusで任意のスキルを実行

mixed

Multiple LLM backends in one workshop. Claude and Codex agents working together. Use when different models have different strengths.

Manusで実行

概要

Multiple LLM backends in one workshop. Claude and Codex agents working together. Use when different models have different strengths.

インストールコマンド

npx skills add https://github.com/joshrotenberg/agent_workshop --skill mixed

このコマンドをClaude Codeにコピー＆ペーストしてスキルをインストール

ソース

joshrotenberg/agent_workshop

スター0

フォーク0

更新日2026年4月2日 19:53

SKILL.md

readonly

name	mixed
description	Multiple LLM backends in one workshop. Claude and Codex agents working together. Use when different models have different strengths.
license	MIT
metadata	{"author":"joshrotenberg","version":"1.0"}

Mixed Backends

Different LLMs for different roles. Same Workshop API regardless of backend.

Setup

configure(backend: AgentWorkshop.Backends.Claude,
  backend_config: ClaudeWrapper.Config.new(working_dir: "."))

agent(:impl, "You write code.", model: "sonnet")
agent(:reviewer, "Review only.",
  backend: AgentWorkshop.Backends.Codex,
  backend_config: CodexWrapper.Config.new(working_dir: "."),
  model: "o3")

When to use

Different models excel at different tasks
You want multiple perspectives from different LLMs
Cost optimization (cheaper model for simple tasks)

このリポジトリの他の Skills

同じリポジトリ

workflow

joshrotenberg/agent_workshop

Declarative multi-stage workflows. Define staged pipelines where each stage feeds its result to the next. Use for structured multi-agent tasks.

2026-04-060

github

joshrotenberg/agent_workshop

GitHub integration via gh CLI. List issues, create PRs, comment on issues. Use for agents that bridge GitHub and the Workshop board.

2026-04-060

board-worker

joshrotenberg/agent_workshop

Autonomous board worker. Receives work items, implements them, reports results. Designed for agents that poll the board and execute tasks independently.

2026-04-050

board

joshrotenberg/agent_workshop

Work board driven development. Post typed work items with dependencies, agents poll by role. Use for multi-step features where ordering matters.

2026-04-020

monitor

joshrotenberg/agent_workshop

Scheduled agent for periodic checks. Use for CI monitoring, PR scanning, health checks, or any polling-based workflow.

2026-04-020

orchestrator

joshrotenberg/agent_workshop

Coordinate a team of AI agents to build software. Break tasks down, create specialist agents from profiles, delegate via ask/cast, review results, and clean up. Use when you need to orchestrate multi-agent workflows.

2026-04-020

ソース

joshrotenberg

joshrotenberg/agent_workshop

GitHub リポジトリを開く Creator のリポジトリを見る

インストールコマンド

ダウンロード

Manusで実行

役立つ用途SOC

ソフトウェア開発者コンピュータ・数学職15-1252L4

name	mixed
description	Multiple LLM backends in one workshop. Claude and Codex agents working together. Use when different models have different strengths.
license	MIT
metadata	{"author":"joshrotenberg","version":"1.0"}

Mixed Backends

Different LLMs for different roles. Same Workshop API regardless of backend.

Setup

configure(backend: AgentWorkshop.Backends.Claude,
  backend_config: ClaudeWrapper.Config.new(working_dir: "."))

agent(:impl, "You write code.", model: "sonnet")
agent(:reviewer, "Review only.",
  backend: AgentWorkshop.Backends.Codex,
  backend_config: CodexWrapper.Config.new(working_dir: "."),
  model: "o3")

When to use

Different models excel at different tasks
You want multiple perspectives from different LLMs
Cost optimization (cheaper model for simple tasks)