用红蓝双队辩论模式深度解答通用问题，以非讨好型人格的理性人视角输出最终报告。 触发条件：用户提出任何值得深度分析的问题，尤其是存在争议、有多个维度、需要权威数据支撑的问题。 提到"红蓝辩论"、"辩证分析"、"正反两面"、"两种观点"、"帮我深度分析"、"有没有反驳意见"、"理性分析一下"、"客观看待"，或提出任何开放性、争议性问题时必须使用本 skill。 即使用户只是提出一个看似简单的问题如"XX 到底值不值得做"或"XX 是对的吗"， 只要涉及价值判断、策略选择、事实真相等复杂情境，也应使用本 skill。 本 skill 不追求讨好用户，拒绝模糊骑墙，双队结论必须互斥，最终裁判整合并输出面向小白的第一人称深度文章。

Use for authorized security testing of desktop thick clients including local storage, update channels, IPC, traffic, and client-side trust boundaries.

Check installed community skills for updates. Shows a diff and requires explicit approval before applying. Use when the user says "check for updates", "update my skills", "anything new for my installed skills", or when invoked from the registry-sync agent.

Draft a DMCA takedown notice, triage one you received, or draft a §512(g) counter-notice. Use when asserting copyright through a §512(c)(3) takedown with the fair-use and perjury gates, when an incoming takedown needs triage into comply / counter / engage / ignore options, or when drafting a §512(g)(3) counter-notice with the consent-to-federal-jurisdiction gate.

Manage matter workspaces — create, list, switch, close, or detach the active matter. Use in multi-client private practice to keep one client's context separate from another, or when a substantive skill needs to know which matter it's working in.

Invention disclosure first-pass screen — novelty, obviousness, §101 eligibility, bar dates, detectability, and strategic value. Use when an invention disclosure comes in and needs triage on whether to pursue a prior-art search and patent counsel review, investigate further, or decline.

Post-closing M&A integration tracker — phased workplan, consent tracking, contract assignment at scale, weekly status reports. Initializes from whatever deal artifacts are available (purchase agreement, deal summary, closing checklist) and connects to deal-context.md and closing-checklist.yaml from the M&A cold-start. Use when user says "integration", "post-close", "post-closing", "consents outstanding", "contract assignment", "integration status", or "what's left on the deal".

Guided customization of your corporate practice profile — change one thing without re-running the whole cold-start interview. Adjust risk posture, escalation contacts, active modules (M&A / Board / Public Company / Entity Management), materiality thresholds, disclosure schedule format, consent precedents, or matter workspace paths. Use when the user says "change my [thing]", "update my profile", "edit my config", or "customize".

Run the cold-start interview — learns your AI governance practice and writes `~/.claude/plugins/config/claude-for-legal/ai-governance-legal/CLAUDE.md` from your AI policy, a reference impact assessment, and key vendor AI agreements. Use when the practice profile is missing or contains `[PLACEHOLDER]` markers, or when user says "set up ai governance plugin", "onboard me", "configure ai governance".

Monthly delta tracking and progress reporting for GEO clients. Compares two GEO audits (baseline vs. current), calculates score improvements across all categories, tracks action item completion, and generates a "here's your progress" client report. Use when user says "compare", "delta", "monthly report", "progress", "confronta", "progressi", "report mensile", or when running a monthly client check-in.

Audits Convex performance for reads, subscriptions, write contention, and function limits. Use for slow features, insights findings, OCC conflicts, or read amplification.

Builds reusable Convex components with isolated tables and app-facing APIs. Use for new components, reusable backend modules, integrations, or component boundary work.

Read or configure the vault filing methodology and suggest destinations for planned knowledge creation under Generic, LYT, PARA, or Zettelkasten. Use for wiki mode, methodology mode, what is my vault mode, set vault mode, switch to PARA, use LYT, Zettelkasten setup, change mode, configure mode, or methodology routing. This skill does not save content or migrate notes.

Detect and use the official Obsidian command-line interface for read-only vault access; use for wiki-cli, Obsidian CLI, Obsidian read, Obsidian search, vault transport, which transport, transport detection, backlinks, tags, or Obsidian command line. Mutations always go through the claude-obsidian transaction core.

Save a user-selected answer, decision, insight, or session summary into an Obsidian vault as one reviewed transaction. Use only when the user explicitly asks to preserve specific conversation content, not when they supply a file or URL to ingest. Triggers: /save, save this, save that answer, file this conversation, save this analysis, keep this insight, preserve this chat result.

Optionally create a draft PR to the upstream repo from a local source directory. Used by coding agents when a local source path is available in the context file.

Performs ARA Seal Level 2 semantic epistemic review on Agent-Native Research Artifacts, scoring six dimensions (evidence relevance, falsifiability, scope calibration, argument coherence, exploration integrity, methodological rigor) and producing a constructive, severity-ranked report with a Strong Accept-to-Reject recommendation. Use after Level 1 structural validation passes, when an ARA needs an objective epistemic critique before publication or release.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques. Use when processing long documents (32k-128k+ tokens), extending pre-trained models beyond original context limits, or implementing efficient positional encodings. Covers rotary embeddings, attention biases, interpolation methods, and extrapolation strategies for LLMs.

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.