LLM 與 Agent

A set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (weekly sprint reports, finance reports, project updates, design news, etc.).

A2A agent card JSON templates with schema validation and examples for different agent types. Use when creating agent cards, implementing A2A protocol discovery, setting up agent metadata, configuring authentication schemes, defining agent capabilities, or when user mentions agent card, agent discovery, A2A metadata, service endpoint configuration, or agent authentication setup.

SQL expert for Rice Data Portal queries using DuckDB SQL. ALWAYS use this skill when data is requested from the Rice stock market database (tickers, prices, fundamentals, insider trades, valuation metrics, etc.). When users request monthly/weekly returns OR momentum, this skill automatically calculates BOTH metrics from end-of-month or end-of-week prices. Returns are expressed as decimals (not percentages). This ensures proper SQL generation, filename prompting, and multi-format file saving. Do NOT use this skill for analyzing local CSV/Parquet files that are already saved.

This skill generates planning documents (기획서) in PowerPoint format from analyzed screen data and project artifacts. It should be used by the auto-draft-orchestrator agent during Phase 4 to create the final PPT output. Not intended for direct user invocation.

このスキルは外部のAI エージェントを実行するためのスキルです。実装や調査の作業を別のAIエージェントに委譲してください。あなた（メインのClaude）はユーザーとの対話・設計判断・方針決定に専念し、実際の調査や実装はこのスキルに任せることでトークンとコンテキストを大幅に節約できます。重要：実装や調査タスクが発生したら、詳細を考える前にまずこのスキルを使ってください。実装の細かいやり取りをメインの会話で行うのはトークンの無駄になります。また、あなたが中間に挟まることで、文脈を省略しがちな人間とタスク実行者との間での口論を防ぐこともできます。

SuperClaude 增强层 Skill，为其他 Skills 提供 SuperClaude 能力增强。支持前置分析、后置审查、智能建议等增强模式。

PMC memory system for persistent context across sessions. Store, search,and retrieve project knowledge using semantic vector search (OpenAI embeddings)and full-text keyword search.Use when: storing important context, searching for past solutions,retrieving project-specific knowledge, or building up institutional memory.

Build complete agent prompts deterministically via Python script. Use BEFORE spawning any BAZINGA agent (Developer, QA, Tech Lead, PM, etc.).

检查VS→GOAL→SC覆盖完整性，确保每个VS都有对应的GOAL和SC。当GOAL和SC创建完成后使用，避免覆盖缺口导致验收失败。

Generate persistent test cases from project analysis, or add individual test cases interactively. Supports full project analysis or adding single test cases via prompt description with browser exploration.

Execute Indirect Prompt Injection attacks for Gray Swan AI Arena Wave 2 with pre-built payloads, model profiling, and evidence collection automation

This skill should be used when the user asks to "find code", "search codebase", "locate functions", "understand architecture", "find implementations", or needs intelligent code discovery using MorphLLM's AI-powered search tools.

将用户输入的简单提示词优化为符合 Nano Banana Pro 最佳实践的高质量提示词。当用户需要为 Gemini 3 Pro Image 生成图像、优化 AI 绘画提示词、或提及 Nano Banana Pro 时使用。

Use this skill to create or update the root AGENTS.md file to register SynthesisFlow skills for AI agent discovery. Triggers include "register SynthesisFlow", "update AGENTS.md", "setup agent guide", or initializing a new project.

Specialized agent for integrating BarqNet backend with multiple client platforms (Desktop/Electron, iOS/Swift, Android/Kotlin). Handles API contracts, authentication flows, error handling, token management, cross-platform compatibility, and end-to-end data flow validation. Use when connecting backend APIs to client applications or debugging integration issues.

Execute TDD red-green-refactor cycle with just verify gate. Use when writing new features or fixing bugs that require test coverage.

Guide for developing, validating, and packaging new Claude skills following the v1.0 specification with automated tooling

File-to-agent mapping and manifest generation for composite monorepo audits. Provides dynamic discovery of all config agents, generates structured audit manifests with priority-ordered spawn instructions, and handles library vs consumer repo exclusions. Use when performing root-level composite audits that scan repository for all config files and coordinate parallel agent execution across 26+ config domains.

Quantum-resistant, self-learning version control for AI agents with ReasoningBank intelligence and multi-agent coordination

Git worktree management with Worktrunk CLI. Use when the user mentions git worktrees, parallel agents, running multiple Claude instances, branching workflows, or wants to create/switch/merge/remove worktrees. Helps with wt switch, wt list, wt merge, wt remove commands and hook configuration.