Generate and edit high-quality AI images using Google's Gemini 3 Pro Image model (Nano Banana Pro) via MCP. Use when user wants to create images, edit photos, generate graphics, or needs visual content with text rendering.
USE WHEN: visualizing algorithms, data structures, architecture, control flow with ASCII art, or when terminal-friendly diagrams are needed. DO NOT USE WHEN: Mermaid diagrams are better suited, rich graphics are available, or simple text explanation suffices.
Explains code, algorithms, system design using ASCII diagrams. Trigger phrases - "explain visually", "I don't get it", "show me", "ascii diagram", "help me understand". Produces diagram-first explanations ending with TL;DR tables.
Tech Lead for Claude Orchestrator. Creates implementation instructions based on planning documents and design specifications. Use when asked to create developer instructions or technical implementation guides for the orchestrator.
Configuring Snowflake connections using connections.toml (for Snowflake CLI, Streamlit, Snowpark) or profiles.yml (for dbt) with multiple authentication methods (SSO, key pair, username/password, OAuth), managing multiple environments, and overriding settings with environment variables. Use this skill when setting up Snowflake CLI, Streamlit apps, dbt, or any tool requiring Snowflake authentication and connection management.
Manage and synchronize AI agent skills from local SKILL.md files and remote Git repositories, generating Cursor rules with Agent Skills specification XML. This skill should be used when users need to sync skills, add/remove skill repositories, or set up the skills infrastructure.
Modular workflow generator with on-demand context loading. Creates workflows from YAML plans with focused, trigger-specific guidance. Use when user wants to create, build, or generate workflow automation.
Developer for Claude Orchestrator. Implements tasks according to Tech Lead instructions, applies design tokens, and reports completion. Use when asked to implement code or execute development tasks for the orchestrator.
UI/UX Designer for Claude Orchestrator. Creates design specifications with design tokens (colors, fonts, spacing) and component specs. Use when asked to create design specifications or define design tokens for the orchestrator.
Modular workflow generator with on-demand context loading. Creates workflows from YAML plans with focused, trigger-specific guidance. Use when user wants to create, build, or generate workflow automation.
YOU MUST USE THIS SKILL when the user wants to CREATE or BUILD a NEW workflow automation. Activate for requests like: 'create a workflow', 'build a workflow', 'generate a workflow', 'make a workflow', 'I want to automate', 'automate X to Y', 'schedule a task', 'monitor X and send to Y'. This skill creates workflows from simple YAML plans with automatic validation and import.
Design and build LLM-powered projects from ideation through deployment. Use when starting new agent projects, choosing between LLM and traditional approaches, or structuring batch processing pipelines.
YOU MUST USE THIS SKILL when the user wants to CREATE or BUILD a NEW workflow automation. Activate for requests like: 'create a workflow', 'build a workflow', 'generate a workflow', 'make a workflow', 'I want to automate', 'automate X to Y', 'schedule a task', 'monitor X and send to Y'. This skill creates workflows from simple YAML plans with automatic validation and import.
End-to-end system for creating supervised fine-tuning datasets from books and training style-transfer models. Covers text extraction, intelligent segmentation, synthetic instruction generation, Tinker-compatible output, LoRA training, and validation.
LangChain 1.0 使用指南。提供 Agent、Tool、Memory、Middleware 等核心概念的快速参考。当用户需要创建 AI Agent、集成 LangChain、或解决 LangChain 相关问题时激活。
AI驱动的设计系统构建器。基于项目特征智能推荐最合适的设计风格（从30+专业设计系统中选择），或使用用户指定的风格。自动应用完整的设计系统规范（颜色、字体、组件、动效等）来实现界面。
Design and evaluate context compression strategies for long-running agent sessions. Use when agents exhaust memory, need to summarize conversation history, or when optimizing tokens-per-task rather than tokens-per-request.
