AI Engineering

Shogun Agent Status displays the operational state of all agents (karō, ashigaru 1-7, strategist) by combining tmux pane status, task YAML status, and unread inbox counts. Use it to check which agents are idle, busy, or missing before assigning tasks.

Shogun Bloom Config asks interactive multiple-choice questions about Claude and ChatGPT subscriptions, then generates a ready-to-paste capability_tiers YAML configuration and recommended fixed agent model assignments for multi-agent routing.

Shogun Model List provides a reference table of AI CLI tools, available models, required subscriptions, and maximum Bloom cognitive level per model. Use it before configuring capability_tiers in multi-agent-shogun.

Shogun Model Switch updates agent CLI type, model, and thinking settings in settings.yaml, then restarts the selected tmux pane. Use it to switch agents between Claude, Codex, Copilot, or Kimi, or toggle extended thinking without leaving the session.

Skill Creator designs, creates, validates, and reviews Claude Code skills (SKILL.md) following the Anthropic 2026-03 guide. Use it for new skill creation, improving existing skills, description quality checks, and trigger test design.

Codex CLI runs OpenAI Codex CLI commands for code analysis, refactoring, and automated editing, with Claude handling model, reasoning effort, sandbox, and resume options. Use when the user asks to run Codex CLI or references OpenAI Codex.

Claude Reflect captures corrections during Claude Code sessions and queues them for review; running /reflect applies them to CLAUDE.md so the agent remembers preferences and project conventions. Use when discussing learnings, corrections, or remembering something for future sessions.

Analyzing Projects examines codebases to understand structure, tech stack, patterns, and conventions, and is used when onboarding to a new project, exploring unfamiliar code, or answering 'how does this work?' questions.

Parallel Execution provides patterns for running multiple subagents concurrently using the Task tool with run_in_background. Use it when coordinating independent tasks, spawning dynamic subagents, or parallelizing feature implementation.

Azure AI provides access to Azure AI Search, Speech, OpenAI, and Document Intelligence services for search, vector/hybrid search, speech-to-text, text-to-speech, transcription, and OCR.

Azure AI Gateway configures Azure API Management as an AI gateway for governing AI models, MCP tools, and agents. Use it for semantic caching, token limits, content safety, load balancing, and AI cost control.

Microsoft Entra Agent ID provisions Microsoft Entra Agent Identity Blueprints, BlueprintPrincipals, and per-instance Agent Identities via Microsoft Graph, and configures OAuth 2.0 token exchange including fmi_path, OBO, and cross-tenant. Use it to give each AI agent a distinct, independently-scoped identity with audit trails and to deploy the Microsoft Entra SDK for AgentID sidecar.