Research

Initialize session as ORCHESTRATOR agent with multi-agent coordination capability. Use for complex tasks requiring parallel agent spawning and result synthesis.

Use when the user needs technical research, architecture validation, documentation generation, or code review. Default is read-only mode, but can enable auto-approve (-y) for code changes.

This skill should be used when the user asks to "搜索文献", "查找学术资料", "搜索最新研究", "literature search", "find papers", "search academic literature", "latest research", or mentions specific academic domains like AI, machine learning, computer science research. Provides comprehensive academic literature search and analysis capabilities for WeChat content creation.

Guide users through TimeGPT fine-tuning workflows - from dataset prep to comparison experiments

Meteor Client addon development for Minecraft. Use when creating, updating, or working with Meteor Client addons - a Fabric mod framework for Minecraft. Covers workspace setup from the template repository, updating addons when Minecraft/Meteor versions change, finding reference implementations from verified addons, and understanding Meteor's addon structure and APIs.

Executive coaching skill that helps you plan your 2026 using research-backed process goals. Guides you through discovery questions, creates outcome goals, converts them to daily behaviors, and sets up accountability systems. Use when you want to plan your year, set goals, or need a coach to help you stay focused.

Run systematic growth experiments to increase acquisition, activation, retention, and revenue. Use when optimizing conversion funnels, running A/B tests, improving metrics, or when users mention growth, experimentation, optimization, or scaling user acquisition.

Use when the user asks about their notes, wants to brainstorm, or references their vault. Provides context about existing notes to inform responses.

Transform experiment workflows into production-ready inference pipelines with orchestration

This skill empowers Claude to construct recommendation systems using collaborative filtering, content-based filtering, or hybrid approaches. It analyzes user preferences, item features, and interaction data to generate personalized recommendations. Use this skill when the user requests to build a recommendation engine, needs help with collaborative filtering, wants to implement content-based filtering, or seeks to rank items based on relevance for a specific user or group of users. It is triggered by requests involving "recommendations", "collaborative filtering", "content-based filtering", "ranking items", or "building a recommender".

Complete reference for Rubin Observatory design system CSS variables and design tokens. Use this skill when styling components, referencing CSS custom properties, understanding the design token architecture, or working with colors, spacing, typography, and other design primitives. Covers both rubin-style-dictionary foundation tokens and squareone-specific tokens.

This skill enhances Claude's ability to conduct web research and translate findings into actionable GitHub issues. It automates the process of extracting key information from web search results and formatting it into a well-structured issue, ready for team action. Use this skill when you need to research a topic and create a corresponding GitHub issue for tracking, collaboration, and task management. Trigger this skill by requesting Claude to "research [topic] and create a ticket" or "find [information] and generate a GitHub issue".

AIシステムへのプロンプトインジェクション攻撃を防ぎ、入力検証とコンテキスト分離の設計指針を提供するスキル。Anchors:• OWASP LLM Top 10 / 適用: LLMセキュリティ脅威モデリング / 目的: インジェクション攻撃の分類と防御パターン理解• Simon Willison's Prompt Injection Research / 適用: 実攻撃パターン分析 / 目的: 実世界の攻撃事例から防御戦略を導出• Defense in Depth principle / 適用: 多層防御設計 / 目的: 単一障害点の排除Trigger:Use when designing prompt injection defenses, implementing AI security measures, sanitizing user inputs for LLM systems, separating trusted and untrusted contexts, conducting security reviews for LLM applications, mitigating indirect prompt injection risks.

Use after research (Z01 files exist) when you need to create implementation plan - wrapper that loads Z01 research context, invokes superpowers:writing-plans, and saves to Z02_{feature}_plan.md in docs/ai/ongoing/

CVE research and security patch workflow for Ark. Provides CVE API integration, mitigation strategies, and security-focused PR templates. Works with research, analysis, and setup skills for comprehensive vulnerability fixing.

Use when the user mentions Exa research OR when the workflow benefits from complex, multi-step research and other exa-ai approaches are not yielding satisfactory results.

This skill enables Claude to design and execute chaos engineering experiments to test system resilience. It is used when the user requests help with failure injection, latency simulation, resource exhaustion testing, or resilience validation. The skill is triggered by discussions of chaos experiments (GameDays), failure injection strategies, resilience testing, and validation of recovery mechanisms like circuit breakers and retry logic. It leverages tools like Chaos Mesh, Gremlin, Toxiproxy, and AWS FIS to simulate real-world failures and assess system behavior.

Assist with core product management activities including writing PRDs, analyzing features, synthesizing user research, planning roadmaps, and communicating product decisions. Use when you need help with PM documentation, analysis, or planning workflows that integrate with your codebase.

Quick Liquid syntax reference and help. Use for fast Liquid questions without full project context loading.

Identify decision-maker contacts at target companies using web research for LinkedIn queries, email patterns, and phone research methods based on ICP buyer personas. Use when you need to find contacts at qualified prospects.