DevOps e Infra

Read historical Agentic SWMM experiment audit artifacts and summarize repeated assumptions, QA issues, failures, missing evidence, run-to-run differences, lessons learned, and controlled skill update proposals. Use downstream of swmm-experiment-audit when multiple audited runs exist or when a user asks for modeling memory, failure-pattern extraction, lessons learned, or human-reviewed skill refine

ActiveRecord patterns, migrations, validations, callbacks, associations

Use after implementation to review a diff or completed phase. A thin ControlFlow layer over Claude Code's native /code-review: adds plan-vs-implementation scope-drift comparison, evidence discipline, and proactive vulnerability/error search. Does not duplicate native review — delegates mechanical/style review to /code-review.

Diagnose and resolve performance issues in Microsoft Fabric Delta Lake and Spark workloads. Use when remediate slow Spark notebooks, long-running Spark jobs, Delta table query degradation, small file problems, VOrder configuration, resource profile tuning, autotune configuration, capacity throttling (HTTP 430), session startup delays, executor memory errors, skewed partitions, or streaming ingesti

Section-by-section media planning and generation. Image generation (GPT Image 1.5 primary, built-in fallback), logo/icon generation (Ideogram v3 → favicon set), video generation (Sora), social preview images (OG 1200x630 + AI search optimization), stock photo curation (Pexels, Pixabay), critique/remix loops (max 3 rounds), asset compression pipeline, and media performance budgets.

Load relevant memories at session start for context continuity. IMPORTANT: Use this AUTOMATICALLY at the START of every session. Search for memories related to: - Current working directory/project - Recent activity (last 7 days) - Topics the user mentions in their first message Execute silently - do not announce what you're loading.

Rewrite a support response in a different tone - formal, casual, technical, empathetic, or concise.

Enhanced mesh coordination with MeshResolver nodes for conflict detection, resolution strategies, and multi-agent file ownership

Decompose complex work into atomic tasks with dependency ordering, priority assignment, and cost estimation using TaskDecomposer nodes

Check health and status of all orchestrated agents

code-forge 상태 대시보드. REFLECT flag, quality 이벤트, notepad/decisions, usage 집계를 한 번에. forge-glow 같은 외부 도구는 --json으로 파싱.

Judge subskill for simmer. Scores a candidate artifact against user-defined criteria on a 1-10 scale and produces ASI (highest-leverage direction) for the next generator round. Supports judge-only, runnable evaluator, and hybrid evaluation modes. Do not invoke directly — dispatched as a subagent by the simmer orchestrator.