自動化

Hugging Face Transformers best practices including model loading, tokenization, fine-tuning workflows, and inference optimization. Use when working with transformer models, fine-tuning LLMs, implementing NLP tasks, or optimizing transformer inference.

Complete Azure DevOps automation - boards, repos, pipelines, artifacts

Graceful degradation through cascading fallback strategies - ensures system always completes while maintaining acceptable functionality

Converts architecture descriptions, module specs, or workflow docs into Mermaid diagrams. Use when visualizing brick module relationships, workflows (DDD, investigation), or system architecture. Supports: flowcharts, sequence diagrams, class diagrams, state machines, entity relationship diagrams, and Gantt charts. Generates valid Mermaid syntax for embedding in markdown docs.

Use when building ML pipelines, orchestrating training workflows, automating model lifecycle, implementing feature stores, or managing experiment tracking systems.

Auto-invokes ultrathink workflow for any work request (default orchestrator)

6-phase investigation workflow for understanding existing systems. Auto-activates for research tasks. Optimized for exploration and understanding, not implementation. Includes parallel agent deployment for efficient deep dives and automatic knowledge capture to prevent repeat investigations.

Use when writing E2E tests with Playwright, setting up test infrastructure, or debugging flaky browser tests. Invoke for browser automation, E2E tests, Page Object Model, test flakiness, visual testing. Keywords: Playwright, E2E, browser testing, automation, Page Object.

Use when writing tests, creating test strategies, or building automation frameworks. Invoke for unit tests, integration tests, E2E, coverage analysis, performance testing, security testing. Keywords: testing, QA, unit test, integration test, E2E, coverage, pytest, Jest.

Initialize a new spec with requirements, design, and task planning. Orchestrates the entire spec-driven workflow from feature idea to actionable task list.

Create and manage Teachable online courses with browser automation and AI content generation. Use when building courses, setting up curriculum, generating content with CCG, or automating Teachable admin tasks.

Use when creating agents. Routes to Simple (30-55 lines), Standard (100-300 lines), or Full (500+ lines) path based on complexity. Triggers: "create agent", "automate workflow". If thinking "just create basic agent" - that's valid for Simple path.

Create new Claude Skills for Miyabi project. Use when building custom skills, packaging workflows, or extending Claude's capabilities. Triggers on "create skill", "make skill", "new skill", "スキル作成", "スキルを作る".

Use when designing deployment pipelines, CI/CD, terraform, or infrastructure automation. Enforces rollback checkpoint then TodoWrite with 19+ items. Triggers: "deploy", "CI/CD", "kubernetes", "terraform". If thinking "rollback later" - use this first.

Systematic error diagnosis and debugging workflow for Rust code. Use when code isn't working, tests fail, or runtime errors occur.

Use when working with Linear tickets/issues - establishes workflows for creating and updating tickets (issues)

Build autonomous coding agents with CLI integration. Use when creating automated coding workflows, managing multi-session development, or building CI/CD code generation pipelines.

Test-Driven Development workflow for Miyabi. Red-Green-Refactor cycle with Rust-specific patterns. Use when implementing new features, fixing bugs, or writing tests.

Agent assignment matrix, blocker escalation, and TDM coordination patterns. Use when assigning work to specialists, managing blockers, or coordinating multi-agent workflows.

Agentic orchestration patterns for long-running tasks. Implements evidence-based delivery and Simon Willison's agent loop. Use when managing multi-step work, coordinating subagents, or orchestrating PR workflows.