SandoDesignSystem

Creates minimal boilerplate structure for new Web Components in Sando Design System following COMPONENT_ARCHITECTURE.md 7-file pattern. Generates only what developer requests - no assumptions, no dead code. Ask first, then scaffold component files, tests, stories based on requirements. <example> User: "Create a new Card component" Assistant: "I'll use the component-creator skill to scaffold the sando-card component with the 7-file pattern." </example> <example> User: "I need a new Accordion component with expand/collapse functionality" Assistant: "I'll use component-creator to scaffold sando-accordion with the specified behavior." </example> <example> User: "Scaffold a new input field component" Assistant: "Using component-creator to generate sando-input with the standard 7-file structure." </example>

Compact guidelines digest for Sando Design System agents. Replaces the full `## 📚 MANDATORY: Read Guidelines Before ANY Work` block that is duplicated across every specialist agent. Provides pre-digested, token- efficient rules organized by agent type. Trigger: inject into sub-agent prompts before task-specific instructions to avoid each agent re-reading full .toon guideline files (saves ~2000-5000 tokens per workflow).

PR creation workflow for Sando Design System. Ensures conventional commits, correct branch naming, CI-passing PRs, and well-structured PR bodies. Trigger: When creating a pull request, opening a PR, or preparing changes for review.

Issue creation workflow for Sando Design System. Creates well-structured GitHub issues for bug reports and feature requests following project conventions. Trigger: When creating a GitHub issue, reporting a bug, or requesting a feature.

Routing table, classification protocol, and ask protocol for the Sando Orchestrator. Extracted from the orchestrator to reduce its context size. Contains the keyword → agent mapping, multi-agent workflow definitions, skill injection reference, and the SDD architectural gate. Trigger: loaded once per session by the orchestrator before any delegation.

Canonical verification protocol for Sando Design System agents. Provides the exact commands, decision trees, and thresholds each specialist agent must run after completing work. Replaces per-agent `<verification>` blocks that are duplicated with slight variations across agents. Trigger: inject into sub-agent prompts so each agent knows exactly what to run and what "done" means.

Sync delta specs to main specs and archive a completed change. Trigger: When the orchestrator launches you to archive a change after implementation and verification.

Create technical design document with architecture decisions and approach. Trigger: When the orchestrator launches you to write or update the technical design for a change.

Explore and investigate ideas before committing to a change. Trigger: When the orchestrator launches you to think through a feature, investigate the codebase, or clarify requirements.

Create a change proposal with intent, scope, and approach. Trigger: When the orchestrator launches you to create or update a proposal for a change.

Write specifications with requirements and scenarios (delta specs for changes). Trigger: When the orchestrator launches you to write or update specs for a change.

Validate that implementation matches specs, design, and tasks. Trigger: When the orchestrator launches you to verify a completed (or partially completed) change.

Step-by-step workflow for creating components in Sando Design System. Ensures no deliverable is forgotten: structure, tokens, logic, tests, stories, changelog, and verification. Each step references specific guidelines for the agent to consult. <example> User: "Create a Card component" Assistant: "I'll follow the component-development-workflow to create sando-card with all deliverables." </example> <example> User: "Build a new Tooltip component" Assistant: "Using component-development-workflow to ensure structure, tokens, tests, stories, and changelog are all completed." </example>

Engram persistent memory protocol for Sando Design System sub-agents. Defines WHEN to save, WHAT to save, HOW to tag observations, and HOW to search memory before starting work. Inject into every sub-agent prompt alongside agent-guidelines-compact and verification-protocol. Trigger: injected by the orchestrator into every specialist sub-agent.

Spec-Driven Development pipeline for architectural changes in Sando Design System. Runs explore → propose → spec → design → tasks phases BEFORE any implementation is delegated. Mode: hybrid (Engram + openspec/ filesystem). Trigger: loaded by the orchestrator when the SDD Architectural Gate fires — i.e., when a change is classified as architectural (breaking change, new pattern, token system restructure, guideline affecting 2+ agents). NOTE: sando-architect IS the domain expert for all SDD phases in this project. Generic sdd-* phase agents are used only for persistence boilerplate.

Break down a change into an implementation task checklist. Trigger: When the orchestrator launches you to create or update the task breakdown for a change.

Create or update the skill registry for the current project. Scans user skills and project conventions, writes .atl/skill-registry.md, and saves to engram if available. Trigger: When user says "update skills", "skill registry", "actualizar skills", "update registry", or after installing/removing skills.

Parallel adversarial review protocol that launches two independent blind judge sub-agents simultaneously to review the same target, synthesizes their findings, applies fixes, and re-judges until both pass or escalates after 2 iterations. Trigger: When user says "judgment day", "judgment-day", "review adversarial", "dual review", "doble review", "juzgar", "que lo juzguen".

Adds skeleton loading states to components by composing sando-skeleton primitives. Creates _renderSkeleton() method that mirrors the component's visual structure. Assumes the calling agent has already determined skeleton is the appropriate loading pattern. <example> User: "Add skeleton loading state to Card component" Assistant: "I'll analyze the Card's layout and create _renderSkeleton() using skeleton primitives." </example> <example> User: "Add skeleton to the Profile component" Assistant: "I'll compose avatar and text skeleton primitives to match the Profile layout." </example>

Optimizes and improves prompts for developers working with LLMs. Analyzes existing prompts, identifies weaknesses, structures loose ideas, and applies professional prompting techniques to maximize results. Use PROACTIVELY when facing complex tasks where optimizing the prompt would improve outcomes. <example> User: "Quiero desarrollar este proyecto pero tengo ideas sueltas, genera un prompt basándote en las siguientes ideas y conceptos." Assistant: "I'll use the prompt-engineer skill to help structure your ideas into a clear, effective prompt." </example> <example> User: "This prompt isn't working well, can you improve it?" Assistant: "I'll use the prompt-engineer skill to analyze and optimize your prompt." </example>