Takahito-Kinouchi
User任務型指揮（Mission Command）による AI 駆動開発フレームワーク — GitHub テンプレート + Claude Code プラグインマーケットプレイス
Categories
Indexed Skills (5)
code-simplifier
Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality. Use when asked to "simplify code", "clean up code", "refactor for clarity", "improve readability", or review recently modified code for elegance. Focuses on project-specific best practices.
mission-command
Explains Mission-Command Development, a delegation doctrine for AI-driven software work adapted from military mission-type command, and maps to the rest of the suite. Use when starting an AI-driven development effort, when deciding how much to delegate to an AI agent, when a human is unsure what to specify versus leave to the AI, or when someone asks "how should I delegate this to the AI", "what's mission-command development", "how much should I let the agent decide", or "am I micromanaging this agent".
roe
Defines Rules of Engagement (ROE), the human-held layer of hard constraints that an AI agent must never cross, and the category of irreversible/tightly-coupled operations that force escalation to a human. Use when defining project guardrails, when an AI agent is unsure whether it's allowed to do something, when someone asks "what are our rules of engagement", "what should never be automated", "is this operation safe for the AI to do on its own", or "should I escalate this".
sop
Defines the Standard Operating Procedure (SOP) — the AI's own default working method for taking a work item from acceptance through implementation to a human-ratified result, including quality gates and escalation triggers. Use when picking up a new task or work item, when defining a default development workflow for an AI agent, when someone asks "what's our SOP", "what are the quality gates", "when should the agent escalate", or "what's the standard path from a ticket to a merged change".
subagent-driven-development
Explains how an AI orchestrator should delegate implementation, sequencing, and verification to subagents rather than doing the work directly, including how to route tasks to models by capability need. Use when implementing a work item, when deciding whether to spawn a subagent, when an orchestrating agent is about to write code or run commands itself, or when someone asks "should I use a subagent for this", "how do I split work across agents", or "which model should handle this step".
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.