End-to-end solution design — system integration, channel orchestration, identity management, observability, and cross-cutting concerns. [EXPLICIT] Use when the user asks to 'design the full solution', 'integrate multiple systems', 'plan API gateway strategy', 'define identity and security architecture', 'set up observability', or mentions C4 containers, BFF, Zero Trust, SLI/SLO, circuit breaker, or migration planning. [EXPLICIT]

Software and technology viability validator — deep forensic analysis of whether proposed software solutions, AI/ML components, and technology choices are viable substance or speculative smoke. Covers service viability, platform viability, methodology viability, tool viability, and vendor assessment for any service type. Use when the user asks to 'validate technology viability', 'detect vaporware', 'verify AI claims', 'assess software maturity', 'check if this tech actually works', or mentions technology due diligence, software validation, AI feasibility, vendor evaluation, or tech-stack viability. This is the devoted software-specific validator — separate and more critical than the multidimensional feasibility analysis. [EXPLICIT]

Structured Socratic debate for resolving ambiguities, trade-offs, and low-confidence decisions. Produces auditable thesis-antithesis-synthesis records with constitutional principle alignment. [EXPLICIT] Trigger: 'debate', 'ambiguity', 'resolve conflict', 'low confidence', 'Socratic'

Security architecture design — threat modeling, zero trust, identity, encryption, compliance. Use when the user asks to 'design security architecture', 'model threats', 'implement zero trust', 'design IAM', 'plan encryption strategy', 'map compliance requirements', or mentions STRIDE, OWASP, OAuth, RBAC, SOC2, ISO27001, PCI-DSS. [EXPLICIT]

This skill should be used when the user asks to schedule Claude Code work, use /loop, create Claude Code cron tasks, list or delete session crons, use CronCreate, CronList, CronDelete, ScheduleWakeup, choose between Routines, Desktop scheduled tasks, GitHub Actions, channels, hooks, /goal, headless claude -p, or cleanup recurring Claude Code tasks.

Elicits, structures, and validates software requirements using user stories, acceptance criteria in Given/When/Then format, and traceability matrices. [EXPLICIT] Transforms vague stakeholder needs into precise, testable specifications. [EXPLICIT] Trigger: 'user stories', 'acceptance criteria', 'requirements'

Retrieval-augmented generation and embeddings: retrieval pipeline (chunking, reranking, context assembly, grounding/citation), embedding-model selection, dense/sparse/hybrid retrieval. [EXPLICIT] Trigger: 'rag', 'retrieval augmented', 'embeddings', 'vector search', 'reranking', 'chunking', 'grounding'.

This skill should be used when the user provides a structured prompt contract, acceptance criteria, runbook, SPEC, or automation brief and asks which Claude primitive should own it: prompt, /goal, scheduled task, API tool runtime, agent loop, explicit command, reusable skill, static plan, dynamic workflow, hook, external event push, independent work, coordinated parallelism, cloud routine, headless run, CI, or human review. It normalizes a fail-closed v2.2 operating packet, selects exactly one primary primitive, and records composed primitives separately.

Routes the full plugin lifecycle — ideate, build, validate, audit, document, deploy, maintain, fix — to the toolkit's pieces (plugin-builder, check.sh/toolkit-hardener, DEPLOY.md, scaffold/gen scripts, LESSONS-LEARNED). [EXPLICIT] Hub for plugin operations and maintenance. Trigger: 'plugin lifecycle', 'plugin ops', 'maintain plugin', 'deploy plugin', 'plugin maintenance', 'how do I ship this plugin'.

This skill should be used when the user asks to 'create a plugin', 'scaffold a Claude Code plugin', 'convert .claude config to a plugin', 'build a plugin marketplace package', 'add plugin skills, agents, hooks, MCP (Model Context Protocol) servers, LSP (Language Server Protocol) servers, monitors, output styles, or settings', 'validate plugin structure', or 'prepare a Claude Code plugin for local testing, community marketplace review, or official marketplace readiness'. It owns plugin-authoring decisions and distinguishes standalone .claude configuration, shareable namespaced plugins, community marketplace submission, and curated official marketplace inclusion.

Performance assessment — load testing, capacity planning, bottleneck analysis, caching, CDN, SLAs. Use when the user asks to 'analyze performance', 'design load tests', 'plan capacity', 'optimize caching', 'configure CDN', 'define SLAs', 'find bottlenecks', or mentions latency, throughput, p95, saturation, cache hit ratio, edge compute. [EXPLICIT]

Author, review, and package Claude Code output styles, including output-styles/*.md files, frontmatter, plugin output styles, skill and artifact boundaries, and coding-instruction preservation.

Production observability and operations: metrics/dashboards and SLO/SLI, alerting design, structured logging architecture, health-check probes, incident response and post-mortems. [EXPLICIT] Trigger: 'observability', 'monitoring', 'alerting', 'logging', 'slo', 'sli', 'incident response', 'post-mortem'.

This skill should be used when the user asks to research a topic with NotebookLM, add sources to a notebook, run cross-notebook or single-notebook queries, synthesize research, expand context from documents, or generate studio artifacts (audio/slides/infographic) from notebook sources.

Mobile app architecture -- native vs cross-platform, offline-first, state management, release management. Use when the user asks to 'design mobile architecture', 'choose between native and cross-platform', 'implement offline-first', 'plan mobile CI/CD', 'optimize app performance', or mentions Flutter, React Native, KMP, MVVM, SwiftUI, Jetpack Compose, or app store deployment. [EXPLICIT]

Routes a reasoning task to the right pattern (Structured Reasoning, Skeleton-of-Thought, Chain-of-Code, Tree of Thought, Committee/Panel) and applies confidence scoring, bias scan, and verify-the-verifier. [EXPLICIT] Meta-skill that operationalizes the metacognition canon. Trigger: 'metacognition', 'reasoning pattern', 'how should I think about', 'tree of thought', 'committee', 'second opinion', 'high-stakes decision'.

This skill should be used when the user asks to 'create diagrams', 'generate Mermaid', 'visualize architecture', 'diagram flows', 'draw a sequence diagram', 'create a C4 diagram', 'add visual diagrams', or mentions diagramming, visualization, flowcharts, sequence diagrams, Mermaid syntax, architecture diagrams, or visual documentation. Use this skill to embed precise, syntactically valid Mermaid diagrams in any discovery deliverable. [EXPLICIT]

Integrar MCP en Claude Code y Agent SDK: .mcp.json, scopes, env-var expansion, auth, permissions, plugin MCP, headless strict config, channels y custom-tool boundary.

Validacion y retry con error feedback especifico; distinguir recuperable de no recuperable; max 2-3 intentos y luego escalar.

Calidad de descripciones de tools como contrato de seleccion; rename y split sobre overloading para evitar misroute.