Loading...
Loading...
Found 251 Skills
Validar prompts dirigidos a agentes de IA (Claude Code, Cursor, Copilot, etc.) contra reglas de redacción efectiva. Calcular un porcentaje de efectividad del prompt y devolver sugerencias de mejora concretas, más una propuesta de prompt reescrito. Cubre verbos no imperativos, lenguaje conversacional, acciones vagas, términos subjetivos, alcance difuso, prohibiciones implícitas, intenciones múltiples y nombres genéricos. Las reglas de detalle técnico (alcance, nombres exactos) se aplican solo a prompts de implementación; en prompts funcionales (user stories, descripciones de comportamiento) se marcan N/A. Usar siempre que el usuario pida validar, revisar, auditar, mejorar, corregir o "pulir" un prompt antes de enviarlo a un agente, o cuando pegue un prompt y pida feedback sobre cómo está redactado.
Design and build custom Claude Code agents with effective descriptions, tool access patterns, and self-documenting prompts. Covers Task tool delegation, model selection, memory limits, and declarative instruction design. Use when: creating custom agents, designing agent descriptions for auto-delegation, troubleshooting agent memory issues, or building agent pipelines.
Help users build effective AI applications. Use when someone is building with LLMs, writing prompts, designing AI features, implementing RAG, creating agents, running evals, or trying to improve AI output quality.
Iterative UI/UX polishing workflow for web applications. The exact prompt and methodology for achieving Stripe-level visual polish through multiple passes.
Vision and multimodal capabilities for Claude including image analysis, PDF processing, and document understanding. Activate for image input, base64 encoding, multiple images, and visual analysis.
Generate images using FAL.ai nanobanana pro. Use when creating product shots, social graphics, brand assets, or any visual content. Integrates with automation system for direct asset generation in Claude Code.
Amazon Bedrock Prompt Management for creating, versioning, and managing prompt templates with variables, multi-variant A/B testing, and flow integration. Use when creating reusable prompt templates, managing prompt versions, implementing A/B testing for prompts, integrating prompts with Bedrock Flows, optimizing prompt engineering, or building production prompt catalogs.
Create, review, and update Prompt and agents and workflows. Covers 5 workflow patterns, agent delegation, Handoffs, Context Engineering. Use for any .agent.md file work or multi-agent system design. Triggers on 'agent workflow', 'create agent', 'ワークフロー設計'.
Generate Ralph-compatible prompts for research, analysis, and planning tasks. Creates prompts with systematic research phases, synthesis requirements, and deliverable specifications. Use when analyzing codebases, creating migration plans, researching technologies, auditing security, or any task requiring investigation before action.
Execute AI image generation with optimal quality. Use when you need to generate images via Replicate API. Triggers on: generate image, create visual, product shot. Outputs generated images for feedback and iteration.
Build, validate, and deploy LLM-as-Judge evaluators for automated quality assessment of LLM pipeline outputs. Use this skill whenever the user wants to: create an automated evaluator for subjective or nuanced failure modes, write a judge prompt for Pass/Fail assessment, split labeled data for judge development, measure judge alignment (TPR/TNR), estimate true success rates with bias correction, or set up CI evaluation pipelines. Also trigger when the user mentions "judge prompt", "automated eval", "LLM evaluator", "grading prompt", "alignment metrics", "true positive rate", or wants to move from manual trace review to automated evaluation. This skill covers the full lifecycle: prompt design → data splitting → iterative refinement → success rate estimation.
Enhanced reasoning patterns via slash commands (/think, /verify, /adversarial, /edge, /compare, /confidence, /budget, /constrain, /json, /flip, /assumptions, /tensions, /analyze, /trade) or natural language ("argue against", "what could break", "show reasoning", "deep review", "meta-prompts", "thinking modes", "second-best approach", "list assumptions", "opposing perspectives").