Question 1

What is MCP?

Accepted Answer

MCP servers give Claude access to external tools: Notion MCP → Read and write Notion pages. GitHub MCP → Create PRs, manage issues. Memory MCP → Persist memory across sessions.

Question 2

Why only the Karo updates dashboard.md

Accepted Answer

1. Single writer: Prevents conflicts by limiting updates to one agent. 2. Information aggregation: The Karo receives all Ashigaru reports, so it has the full picture. 3. Consistency: All updates pass through a single quality gate. 4. No interruptions: If the Shogun updated it, it could interrupt the Lord's input.

Question 3

Why Mailbox System?

Accepted Answer

Using YAML files as mailboxes provides persistence across crashes, event‑driven notification (inotifywait) with zero CPU while idle, avoids cross‑talk, simplifies debugging (open the file to see message history), serializes writes with flock, and ensures delivery even if agents restart.

Question 4

Why a hierarchy (Shogun → Karo → Ashigaru)?

Accepted Answer

1. Instant response: The Shogun delegates immediately, returning control to you. 2. Parallel execution: The Karo distributes to multiple Ashigaru simultaneously. 3. Single responsibility: Each role is clearly separated. 4. Scalability: Adding more Ashigaru doesn't break the structure. 5. Fault isolation: One Ashigaru failing doesn't affect the others. 6. Unified reporting: Only the Shogun communicates with you.

Question 5

What install.bat does automatically:

Accepted Answer

✅ Checks if WSL2 is installed (guides you if not) ✅ Checks if Ubuntu is installed (guides you if not) ✅ Shows next steps (how to run first_setup.sh).

Question 6

What shutsujin_departure.sh does:

Accepted Answer

✅ Creates tmux sessions (shogun + multiagent) ✅ Launches each agent with the CLI configured in config/settings.yaml (Claude/Codex/Copilot/Kimi/OpenCode) ✅ Auto‑loads instruction files or generated agent definitions for each CLI ✅ Resets queue files for a fresh state ✅ Starts ntfy listener for phone notifications (if configured). After running, all agents are ready to receive commands.

Question 7

Why CLI (Not API)?

Accepted Answer

CLI subscriptions flip the cost model from per‑token API pricing to a flat‑rate monthly fee. Running 8 Opus‑grade agents via the API can cost $100+/hour, while a CLI subscription is around $200/month, giving cost predictability, unlimited usage, and freedom to experiment without token anxiety.

Question 8

What is this?

Accepted Answer

multi-agent-shogun is a system that runs multiple AI coding CLI instances simultaneously, orchestrating them like a feudal Japanese army. Supports Claude Code, OpenAI Codex, GitHub Copilot, Kimi Code, OpenCode, Cursor, and Antigravity. One command spawns 7 AI workers + 1 strategist executing in parallel. AI remembers your preferences across sessions (Memory MCP). Real-time progress on a dashboard.

Question 9

What makes this different

Accepted Answer

Zero coordination overhead — Agents talk through YAML files on disk. The only API calls are for actual work, not orchestration. Full transparency — Every agent runs in a visible tmux pane. Every instruction, report, and decision is a plain YAML file you can read, diff, and version-control. No black boxes. Battle‑tested hierarchy — The Shogun → Karo → Ashigaru chain of command prevents conflicts by design.

Question 10

Why Shogun?

Accepted Answer

Most multi-agent frameworks burn API tokens on coordination. Shogun doesn't. It uses a feudal hierarchy via tmux with zero coordination overhead, allowing 8 independent agents with only the work-related API calls.

Shogun Model Switch (Grade A)Sicherheitsgetestete Daten- und KI-Fähigkeit für Claude AI. Klasse A. Agenten-CLI/Modell wechseln Sie live. settings.yaml aktualisiert → /Ausgang → Neuer CLI startet → Pane-Metadata aktualisiert. Überlegung möglich, ob zugegriffen.

Übersicht

Hauptfunktionen

Preise

Anwendungsfälle

Modellwechsel von Agenten

Batch-Update von Agenten

Pro & Contra

Pro

Contra

Bewertungen

Fragen & Antworten

Frage stellen

Alternativen zu Fähigkeiten

Manage Headers (Grade A)

Using Git Worktrees (Grade A)

Ga4 Bigquery Schema (Grade A)

Meta Capi (Grade A)

Callees (Grade A)

Test Module Name (Grade A)

Board Of Directors (Grade A)

Advpl Mvc Avancado (Grade A)