AgentWorkshop (AgentWorkshop v0.3.0)

Multi-agent orchestration for IEx. Backend-agnostic, MCP-enabled.

AgentWorkshop lets you run multiple LLM agents side by side, coordinating them with simple IEx commands. It works with any CLI-based LLM backend (Claude, Codex, etc.) through a pluggable backend behaviour.

Quick start

import AgentWorkshop.Workshop

configure(
  backend: AgentWorkshop.Backends.Claude,
  backend_config: ClaudeWrapper.Config.new(working_dir: "."),
  model: "sonnet",
  context: "Elixir project. Run mix test before committing."
)

agent(:impl, "You write clean code.", max_turns: 15)
agent(:reviewer, "Review only.", model: "opus")

ask(:impl, "Implement caching")
|> pipe(:reviewer, "Review for edge cases")

Mixed backends

agent(:claude_impl, "You write code.", backend: AgentWorkshop.Backends.Claude)
agent(:codex_reviewer, "Review only.", backend: AgentWorkshop.Backends.Codex)

Backends

• AgentWorkshop.Backends.Claude - Claude Code CLI (requires claude_wrapper)
• AgentWorkshop.Backends.Codex - OpenAI Codex CLI (requires codex_wrapper)

See AgentWorkshop.Backend to implement your own.