</>MCP Agents Market
Agent

multi-agent-shogun

by yohey-w1.4kShellUpdated 2026-08-06

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

Claude CodeOpenAI CodexGitHub CopilotKimi CodeOpenCodeCursor

multi-agent-shogun is a samurai-inspired multi-agent orchestration system that runs up to 10 AI coding assistants simultaneously through a feudal hierarchy (Shogun → Karo → Ashigaru → Gunshi). It uses tmux sessions and YAML file-based messaging to coordinate parallel execution of complex tasks across multiple CLI tools (Claude Code, OpenAI Codex, GitHub Copilot, Kimi Code, OpenCode, Cursor, Antigravity) with zero API coordination overhead. The system enables developers to delegate work that gets automatically broken down and executed in parallel, with real-time progress tracking via dashboard and mobile notifications.

Key Features

Parallel execution of up to 8 AI agents simultaneously with zero coordination API calls
Hierarchical task delegation through Shogun (commander) → Karo (manager) → Ashigaru (workers) + Gunshi (strategist)
Multi-CLI support: mix Claude Code, OpenAI Codex, GitHub Copilot, Kimi Code, OpenCode, Cursor, and Antigravity in one formation
Event-driven YAML-based messaging with inotifywait for zero-polling communication between agents
Cross-session memory persistence via Memory MCP for long-term preference learning
Bottom-up skill discovery where agents propose reusable patterns from actual work
Mobile app for Android with voice input, 9-pane monitoring, and bidirectional ntfy push notifications
Dynamic model routing based on Bloom's Taxonomy cognitive complexity levels (L1-L6)

Use Cases

  • 01Research sprints where 5+ agents simultaneously research different frameworks and compile comparison reports
  • 02PoC preparation with parallel document analysis, technical feasibility checks, and draft creation
  • 03Complex coding tasks broken into subtasks executed by multiple specialized agents in parallel
  • 04Multi-project white-collar work coordination with persistent context across sessions
  • 05Voice-driven task management via phone (SayTask) with AI auto-categorization and streak tracking
  • 06Cost-optimized AI development using flat-rate CLI subscriptions instead of per-token API billing

Related Agents

View more

hermes-agent

The agent that grows with you

Agent250.4k

agency-agents

A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.

Agent155.5k

openinterpreter

A coding agent for open models like Kimi K3

Agent68k

cline

Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Agent66.3k

multi-agent-shogun — FAQ

What is multi-agent-shogun and how does it work?+

multi-agent-shogun is a hierarchical multi-agent system that coordinates multiple AI coding CLI tools (Claude Code, Codex, Copilot, etc.) through tmux sessions and YAML file messaging. You give commands to the Shogun, which delegates to a Karo manager that distributes work to 7 Ashigaru workers and 1 Gunshi strategist for parallel execution.

How do I install and set up multi-agent-shogun?+

On Windows, download and run install.bat to set up WSL2/Ubuntu, then run first_setup.sh in Ubuntu to install dependencies (tmux, Node.js, Claude Code CLI, Memory MCP). Daily startup is via shutsujin_departure.sh which creates tmux sessions and launches all agents. On Linux/macOS, clone the repo and run first_setup.sh directly.

Which AI clients does multi-agent-shogun work with?+

It supports Claude Code, OpenAI Codex, GitHub Copilot, Kimi Code, OpenCode, Cursor, and Antigravity CLI tools. You can mix different CLIs in one formation by configuring settings.yaml to assign each agent a specific CLI type and model.

What are the prerequisites and do I need API keys?+

You need tmux, bash 4+, Node.js v20+, and at least one AI coding CLI installed and authenticated. Windows users need WSL2. Authentication is per CLI (OAuth for Claude Code, API keys for others). The system uses CLI subscriptions instead of API tokens to avoid per-token costs.

Is multi-agent-shogun free to use?+

The multi-agent-shogun system itself is free and open-source (MIT license). However, you need subscriptions or access to the underlying AI CLI tools (Claude Code, Codex, Copilot, etc.). The design optimizes for flat-rate CLI subscriptions over per-token API billing.

How do I connect to and control the agents?+

After running shutsujin_departure.sh, connect to the Shogun via 'tmux attach-session -t shogun' and give commands in natural language. Progress appears in dashboard.md, and you can view all 9 agent panes via 'tmux attach-session -t multiagent'. An Android app provides mobile monitoring and voice control.

How do I install multi-agent-shogun?+

Open the source repository on GitHub and follow its README. multi-agent-shogun is a agent — MCP Agents Market links you directly to the official repo.

Is multi-agent-shogun free?+

multi-agent-shogun is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches

Developer Tools AgentsProductivity & Workflow AgentsPlanning AgentsBest MCP ServersBest SkillsBest PluginsTop rated listingsClaude Code toolsBrowse all categories