yohey-w/multi-agent-shogun

Samurai-inspired multi-agent system for Claude Code. Orchestrate parallel AI tasks via tmux with shogun → karo → ashigaru hierarchy.

What it solves

It eliminates the high API costs and coordination overhead associated with running multiple AI coding agents. Instead of paying per token for orchestration, it leverages flat-rate CLI subscriptions to run several independent agents in parallel without burning tokens on internal communication.

How it works

The system uses a feudal Japanese hierarchy (Shogun $\rightarrow$ Karo $\rightarrow$ Ashigaru) implemented via tmux split-panes and YAML files on disk.

  • Shogun: Receives user commands and delegates them.
  • Karo: A manager agent that breaks down tasks and distributes them to workers.
  • Ashigaru: Seven worker agents that execute tasks in parallel.
  • Gunshi: A strategist agent that reviews reports.

Agents communicate asynchronously through YAML files, meaning no API calls are wasted on coordination. It supports a variety of CLI tools including Claude Code, OpenAI Codex, GitHub Copilot, Kimi Code, OpenCode, Cursor, and Antigravity, using a unified instruction build system to keep rules consistent across different vendors.

Who it’s for

Developers who want to deploy a "army" of AI agents to handle complex coding tasks in parallel while maintaining full transparency and predictable monthly costs.

Highlights

  • Zero Coordination Overhead: Uses disk-based YAML communication instead of API-based orchestration.
  • Multi-CLI Support: Compatible with 7 different AI coding CLIs, allowing users to mix and match strengths.
  • Bottom-Up Skill Discovery: Agents automatically identify reusable patterns during work and propose them as permanent skills for the user to approve.
  • Live Observability: Every agent runs in its own visible tmux pane with a real-time dashboard.
  • Mobile Command Center: Includes a dedicated Android app for monitoring agents and sending commands via SSH and voice input.

Related

  • Project
    awslabs/cli-agent-orchestrator

    A coordination framework that manages multiple AI coding CLIs, allowing a supervisor agent to delegate tasks to specialist workers in parallel or sequence.

  • Project
    asheshgoplani/agent-deck

    Agent Deck is a Go‑written terminal UI that lets you launch, monitor, fork, and orchestrate multiple AI coding agents (Claude, OpenCode, Codex, Pi, …) from a single pane. It adds group‑based policies, per‑group Claude credentials, MCP tool toggles, cost dashboards, git‑worktree handling, and a phone‑controlled *conductor* that can be queried via Telegram/Slack/Discord. Install with a one‑liner script, Homebrew, or `go install`, then manage sessions with keyboard shortcuts (`Enter` to attach, `f` to fork, `$` for cost, `/` to search, etc.).

  • Project
    wanikua/danghuangshang

    A multi-agent collaboration system based on the OpenClaw framework that organizes AI agents into a hierarchical government or corporate structure for efficient task delegation and execution.

  • Project
    777genius/agent-teams-ai

    Agent Teams AI is a free cross‑platform desktop app that lets you build and monitor teams of AI agents (Claude Code, Codex, OpenCode, Cursor, SuperGrok, GitHub Copilot, Z.AI, MiniMax, Kiro, etc.). Agents can create tasks, communicate, run code and terminal commands, and perform code reviews while you watch a Kanban board. The app tracks token usage, budgets, and per‑agent resource stats, and offers a built‑in terminal, Git‑aware editor, multilingual UI, and plug‑in support via an `mcp-server` component.

  • Project
    YoanWai/agent-manager

    A terminal-based manager for AI coding agents that consolidates multiple CLI agents into a single interface with live status tracking and integrated diff reviews.