将多个 PowerPoint 文件合并为一个 PPT。当用户提到 PPT 合并、合并幻灯片、合并 PPTX 时使用。

将 PDF 文件转换为 Word 文档（.docx），支持单文件和批量转换。当用户提到 PDF 转 Word、PDF 转 docx 时使用。

批量修改文件/文件夹名称，替换或删除指定内容。当用户提到批量重命名、批量改名、文件替换、文件名替换时使用。

Install a pasted SKILL.md (or one fetched from a URL or file path) into the user-skill directory, reviewing risk and connector requirements.

File a locally-captured, already-redacted Gini crash report as a GitHub issue, with the user's consent. Reads the pending crash queue and delegates the actual filing to the github-issues skill.

Manage Apple Notes via memo CLI: create, search, edit.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Write publication-ready ML/AI/Systems papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM, OSDI, NSDI, ASPLOS, SOSP. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, reviewer guidelines, and citation verification workflows.

Use when answering a question that needs data from a ktx-connected database - investigating, analyzing, "how many", "show me", "what's the breakdown of", finding records by value, exploring tables, comparing periods, explaining metrics, or any data-analysis request. Triggers even when the user does not say "analytics"; if the answer requires querying a configured ktx connection, this skill applies.

Use for location-first, no-human, and scene-bible workflows that need stable environment naming, spatial continuity, lighting logic, and clean world extraction without character contamination.

Use for screenplay, scene writing, dialogue formatting, beat outlines, and any request that should produce structured script text compatible with Fountain syntax.

Use for fights, chases, transformations, and high-kinetic sequences that need readable geography, motivated camera logic, impact rhythm, and continuity-safe action coverage.

Activate when the user needs to interact with any website — browser automation, web scraping, screenshots, form filling, UI testing, monitoring, or building AI agents. Provides pre-verified page actions with step-by-step instructions and tested selectors.

Facilitate weekly review process with reflection, goal alignment, and planning. Create review notes, analyze past week, plan next week. Use on Sundays or whenever doing weekly planning.

Manage software complexity through deep modules, information hiding, and strategic programming. Use when the user mentions "module design", "API too complex", "shallow class", "complexity budget", "strategic vs tactical", "deep module", "information leakage", "pass-through method", "this code is over-engineered", or "simplify this design". Also trigger when reviewing an interface for simplicity, evaluating whether an abstraction is pulling its weight, deciding whether a comment is worth writing, or choosing between general-purpose and special-purpose approaches. Covers deep vs shallow modules, red flags for complexity, and comments as design documentation. For code quality, see clean-code. For architecture boundaries, see clean-architecture.

Apply meta-principles of software craftsmanship: DRY, orthogonality, tracer bullets, and design by contract. Use when the user mentions "best practices", "pragmatic approach", "broken windows", "tracer bullet", "software craftsmanship", "avoid technical debt", "code ownership", or "how do I become a better developer". Also trigger when evaluating build-vs-buy decisions, designing estimation approaches, or choosing between reversible and irreversible architectural decisions. Covers estimation, domain languages, and reversibility. For code-level quality, see clean-code. For refactoring techniques, see refactoring-patterns.

Manage software complexity through deep modules, information hiding, and strategic programming. Use when the user mentions "module design", "API too complex", "shallow class", "complexity budget", "strategic vs tactical", "deep module", "information leakage", "pass-through method", "this code is over-engineered", or "simplify this design". Also trigger when reviewing an interface for simplicity, evaluating whether an abstraction is pulling its weight, deciding whether a comment is worth writing, or choosing between general-purpose and special-purpose approaches. Covers deep vs shallow modules, red flags for complexity, and comments as design documentation. For code quality, see clean-code. For architecture boundaries, see clean-architecture.

Apply meta-principles of software craftsmanship: DRY, orthogonality, tracer bullets, and design by contract. Use when the user mentions "best practices", "pragmatic approach", "broken windows", "tracer bullet", "software craftsmanship", "avoid technical debt", "code ownership", or "how do I become a better developer". Also trigger when evaluating build-vs-buy decisions, designing estimation approaches, or choosing between reversible and irreversible architectural decisions. Covers estimation, domain languages, and reversibility. For code-level quality, see clean-code. For refactoring techniques, see refactoring-patterns.

仅在编写/更新/修复 C++ 测试、配置 GoogleTest/CTest、诊断失败或不稳定的测试、或添加覆盖率/检测器（Sanitizers）时使用。

Spring Boot 中用于实体设计、关联关系、查询优化、事务、审计、索引、分页和连接池的 JPA/Hibernate 模式。