Root-cause a reconciliation break to its source transaction or posting — follow the audit trail from the break row back to the originating entry on each side and state what differs and why. Use after gl-recon has classified a break.

Identify the precise location of bugs in source code, modules, and systems. Use this skill when debugging applications, investigating test failures, analyzing error reports, tracing runtime issues, or performing root cause analysis. Analyzes stack traces, error messages, failing tests, and code patterns to pinpoint buggy functions, classes, files, or modules with confidence rankings and supporting evidence.

Build a throwaway prototype to flush out a design before committing. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".

Claude Codeスキル（SKILL.md）の設計・作成・バリデーション・レビュー。 Anthropic公式ガイド(2026-03)準拠。新規スキル作成、既存スキルの改善、 description品質チェック、トリガーテスト設計に使用。 「スキル作って」「スキル設計」「SKILL.md作成」「スキルレビュー」で起動。 Do NOT use for: スキルの実行・呼び出し（それは各スキル自体が行う）。

Interactive wizard: guided questions with multiple-choice options about subscriptions, then outputs a ready-to-paste capability_tiers YAML + fixed agent model assignments. Trigger: "capability_tiers", "bloom config", "routing setup", "set up model routing", "ルーティング設定", "capability_tiers設定", "モデル設定", "サブスク設定", "model routing"

Migrate bUnit test files from deprecated beta API (1.0.0-beta-10) to bUnit 2.x stable API. Use this when working on .razor test files in BlazorWebFormsComponents.Test that contain old patterns like TestComponentBase, Fixture, or SnapshotTest.

Official Rails documentation. Use when asked about any Rails-specific topic including ActiveRecord, routing, controllers, views, mailers, jobs, Action Cable, Action Text, Active Storage, migrations, validations, callbacks, associations, caching, security, or internals.

Generate Harness v0 Template YAML for reusable pipeline components (Step, Stage, Pipeline, StepGroup) and create them via MCP. Use when asked to create a template, build a step template, stage template, pipeline template, or reusable step. Do NOT use for checking template usage or references (use template-usage instead). Trigger phrases: create template, step template, stage template, pipeline template, reusable step, template YAML.

Use this skill whenever a user wants to win AI citations on prompts that competitors currently dominate — whether they say "competitors are getting cited instead of us", "we're losing on these prompts", "how do I outrank [competitor] in AI answers", "find prompts where we should be winning", "create content to beat [competitor]", or any variation where the goal is capturing AI share on prompts a competitor currently owns. This skill pulls competitor visibility data from AI Visibility, identifies the specific prompts where competitors win and Amplitude is absent, clusters them by intent, and produces targeted comparison pages, alternatives content, or rebuttal assets — then pushes drafts to CMS. Trigger on any mention of competitor, prompt hijack, outrank, or "why is [competitor] getting cited instead of us".

Plan and produce video and visual assets for product launches — demo videos, GIFs, screenshots, and programmatic video with Remotion. Use this skill whenever someone asks to create a demo video, record a product walkthrough, make a launch GIF, plan visual assets for a launch, or decide what media to create for an announcement. Also trigger for "what should our demo video show," "how do we record the product," "I need visuals for the launch," "make a video for this launch," or any request about creating non-text media tied to a product release. Covers video scripting, recording, programmatic video (Remotion), GIF creation, screenshot best practices, and per-platform format guidance.

This skill should be used when the user says "execute plan with teams", "run plan teams", "execute with agent teams", "team execution", "execute plan with agent teams instead of subagents", "teams mode", "agent teams mode", "execute with teams", or wants to execute a structured project plan using Claude Code's experimental Agent Teams feature. Creates a team of executor, reviewer, and architect teammates that collaborate on task implementation with built-in quality gates. Requires the experimental Agent Teams feature to be enabled. For standard subagent-based execution, use arn-code-execute-plan instead.

Use when porting external PyTorch object detection models in DeepDetect through the external PyTorch worker backend.

按指定页码范围拆分 PDF 文件。当用户提到 PDF 拆分、PDF 分割、PDF 切分、拆 PDF 时使用。

Create animation-rich browser-based HTML presentations from scratch, enhance existing HTML decks, or convert PowerPoint files into web presentations. Use only when the requested deliverable is HTML or a browser presentation; do not use when the user needs an editable native .pptx file.

AI(ChatGPT·Claude·Gemini)가 쓴 한글 텍스트를 사람이 쓴 글처럼 윤문한다. 번역투·영어 인용 과다·기계적 병렬·관용구·피동 남용·접속사 남발·리듬 균일·이모지/불릿 과다 등 10대 카테고리 70개 AI 티 패턴을 탐지·분류해 내용은 한 글자도 건드리지 않고 문체·리듬·표현만 자연스럽게 재작성한다. 트리거 — "AI 티 없애줘", "AI 윤문", "ChatGPT 티 제거", "번역투 고쳐", "사람이 쓴 것처럼", "humanize Korean". 단순 맞춤법 교정·번역·내용 추가는 대상 아님.

Web2 recon pipeline — subdomain enumeration (subfinder, Chaos API, assetfinder), live host discovery (dnsx, httpx), URL crawling (katana, waybackurls, gau), directory fuzzing (ffuf), JS analysis (LinkFinder, SecretFinder), continuous monitoring (new subdomain alerts, JS change detection, GitHub commit watch). Use when starting recon on any web2 target or when asked about asset discovery, subdomain enum, or attack surface mapping.

Bug bounty report writing for H1/Bugcrowd/Intigriti/Immunefi — report templates, human tone guidelines, impact-first writing, CVSS 3.1 scoring, title formula, impact statement formula, severity decision guide, downgrade counters, pre-submit checklist. Use after validating a finding and before submitting. Never use "could potentially" — prove it or don't report.