區塊鏈

Conducts structured interviews to produce Change Contracts - definitions of done expressed in terms of behavior, constraints, operability, and acceptance criteria. Use when externalizing intent for work to be delegated to AI agents, when defining requirements without implementation details, or when asked to create a change contract.

Define and structure database models with proper data types, relationships, constraints, and validation following ORM best practices. Use this skill when creating or modifying database model classes, ORM entities, schema definitions, or data model files. Use this skill when working with SQLAlchemy models, Django models, ActiveRecord models, Prisma schemas, TypeORM entities, or Sequelize models. Use this skill when defining table schemas, establishing relationships (one-to-many, many-to-many), implementing model validation, adding timestamps and soft deletes, or configuring cascade behaviors. Use this skill when working with files in models/, entities/, schemas/, or orm/ directories that define database table structures and business logic.

Efficiently compose multiple MCP tool operations by launching a subagent that writes and executes TypeScript or Python code. Use when you need to compose 3+ MCP tool calls, process their results together, implement complex logic, or avoid token overhead from loading many MCP tool schemas into main context. The subagent has MCP servers configured and writes code that calls tools dynamically.

Create traceable specifications from architecture. Use after architecture is approved to define invariants, edge cases, and acceptance criteria.

Enforce Definition of Done checks from QUALITY_GATES.md.

Clerk and Supabase integration patterns for user sync, JWT authentication, and RLS policies. Use when integrating Clerk authentication with Supabase, syncing user data between platforms, configuring RLS with Clerk JWT tokens, setting up webhooks for user events, implementing secure database access with Clerk identity, or when user mentions Clerk Supabase sync, user synchronization, JWT RLS, authentication webhooks, or database user management.

Fantasy-themed color tokens and semantic color usage. Use when applying colors, creating themes, or ensuring color accessibility. Covers surface/text relationships, state colors, and dark theme patterns. (project)

Encrypt and decrypt BSV backup files (.bep format) using the bitcoin-backup CLI. Supports all backup types (BAP identity, wallet, ordinals, vault). Uses Flow's FLOW_BACKUP_PASSPHRASE and stores files in /.flow/.bsv/ following Flow's BSV convention.

Fix symbol selection failing due to asset-type filter mismatch. Trigger when: (1) '0/N passed hard filter' error, (2) crypto selection fails unexpectedly, (3) assets filtered by wrong price/volume thresholds.

Complete Figma design extraction for pixel-perfect implementation.Extracts design tokens, component specs, layouts, and responsive behavior systematically.Use when user provides Figma design:- Figma URL or file link provided- User mentions "Figma", "design file", "mockup", or "design system"- During /create-plan phase when design needs extraction- User says "extract from Figma" or references Figma linkExtract systematically:- Design tokens: ALL colors (hex + usage), typography (sizes, weights, line heights), spacing scale, border radius, shadows, opacity values- Components: ALL states (default, hover, active, focus, disabled, loading, error), ALL variants (size/style/intent), exact dimensions, spacing, visual properties- Layouts: page structure, grid systems (columns, gutters), component hierarchy- Responsive: mobile/tablet/desktop differences explicitly documented- Assets: icons (names, sizes), images (dimensions, alt text), illustrationsGoal: Extract ONCE completely during planning phase

Premium dark mode design system for NGX Genesis. Use when creating web interfaces, landing pages, dashboards, React components, or any UI for the NGX ecosystem. Includes color tokens, glassmorphism effects, neon glow buttons, depth cards, and all brand visual effects. The style is 80% effects and 20% colors - requires transparencies, backdrop-blur, multiple shadows, and subtle gradients.

Guide developers through property-based testing including property definition, shrinking, and framework-specific implementation

Standardizing and formatting datasets for Unsloth, including chat template conversion and synthetic data generation (triggers: chat templates, ShareGPT, Alpaca, conversation_extension, add_new_tokens, standardize_sharegpt, formatting_prompts_func).

Implements Firebase Authentication with email, OAuth, phone auth, and custom tokens. Use when building apps with Firebase, needing flexible auth methods, or integrating with Firebase ecosystem.

Build AI agents using prompt-native architecture where features are defined in prompts, not code. Use when creating autonomous agents, designing MCP servers, implementing self-modifying systems, or adopting the "trust the agent's intelligence" philosophy.

Ability to design, analyse, test, and maintain mechanical systems and components that meet defined functional, safety, and performance requirements. Includes applying engineering principles to materials, structures, thermodynamics, fluid mechanics, and motion systems; producing and interpreting technical designs and specifications; validating designs through analysis and testing; and supporting manufacture, operation, and lifecycle management. Applies across industrial, infrastructure, energy, manufacturing, and product contexts and is independent of specific tools or industries, with human accountability retained for safety, compliance, and outcomes.

Use when building features with unknowns or uncertainties - start from well-understood parts (business logic, UX, pure functions), write code as if ideal collaborators exist, use mocks to define APIs through usage, preventing analysis paralysis from trying to understand everything upfront

Token-efficient codebase navigation through intelligent symbol loading and querying. Use this skill when implementing new features (find existing patterns), exploring codebase structure, searching for components/functions/types, or understanding architectural layers. Reduces token usage by 60-95% compared to loading full files. Layer split enables 50-80% additional backend efficiency.

Complete visual design language for Arcanea - cosmic theme tokens, component patterns, animation standards, and Academy-specific aesthetics

This skill provides comprehensive knowledge for working with the Anthropic Claude Agent SDK. It should be used when building autonomous AI agents, creating multi-step reasoning workflows, orchestrating specialized subagents, integrating custom tools and MCP servers, or implementing production-ready agentic systems with Claude Code's capabilities.Use when building coding agents, SRE systems, security auditors, incident responders, code review bots, or any autonomous system that requires programmatic interaction with Claude Code CLI, persistent sessions, tool orchestration, and fine-grained permission control.Keywords: claude agent sdk, @anthropic-ai/claude-agent-sdk, query(), createSdkMcpServer, AgentDefinition, tool(), claude subagents, mcp servers, autonomous agents, agentic loops, session management, permissionMode, canUseTool, multi-agent orchestration, settingSources, CLI not found, context length exceeded