Skill UI

A comprehensive, multi-phase workflow designed for Infrastructure as Code (IaC) using Terraform. It guides users through provisioning cloud resources, creating reusable modules, managing complex state, handling multi-environment deployments, and integrating robust CI/CD pipelines and security practices. Ideal for establishing scalable, repeatable, and auditable cloud infrastructure.

A comprehensive collection of production-ready, reusable Terraform modules designed for standardized infrastructure provisioning across major cloud platforms: AWS, Azure, and GCP. This library enforces Infrastructure as Code (IaC) best practices, allowing users to build, test, and deploy multi-cloud compatible components efficiently and reliably.

Provides Terraform and OpenTofu best-practice guidance for module design, testing selection, CI/CD pipeline setup, security scanning, and architecture decisions in production infrastructure-as-code projects.

This skill provides advanced expertise in Infrastructure as Code (IaC) using Terraform and OpenTofu. It covers designing robust, reusable modules, managing complex multi-cloud state backends, and implementing enterprise-grade CI/CD pipelines. It specializes in governance, Policy-as-Code, and achieving GitOps compliance for full infrastructure automation.

A specialized skill designed to ensure robust software quality by covering the entire testing lifecycle. This includes defining test strategies, writing unit, integration, and E2E tests, and implementing advanced testing types like performance (k6), security (OWASP), and functional testing. It guides users through coverage analysis, debugging flaky tests, and generating detailed defect and quality reports.

Document outlines testing strategies such as contract, snapshot, mutation, property-based, integration with containers, and disciplined test organization to help teams design reliable suites that cover behavior, API contracts, and invariants.

This skill maps identified security threats to appropriate controls and mitigation strategies. It is essential for prioritizing security investments, building comprehensive remediation roadmaps, validating existing control coverage, and performing detailed risk treatment planning. It helps organizations design robust, defense-in-depth security architectures.

A specialized skill for conducting comprehensive security architecture reviews, risk assessments, and threat identification. It masters advanced methodologies like STRIDE, PASTA, and attack tree construction. Use this to build secure-by-design systems, identify potential attack vectors, and strengthen system resilience before deployment.

This comprehensive skill provides advanced architectural patterns for implementing robust API throttling and rate limiting policies. It covers critical concepts like concurrency limiting, priority queuing, circuit breakers, and adaptive throttling that adjust limits based on real-time backend health metrics (p95/p99 latency). Use this when designing resilient, scalable backend services that must protect themselves from overload and maintain stability under high load.

This skill enforces a rigorous, spec-driven development lifecycle spanning planning (Specify/Design), task breakdown, execution, and verification. It automatically adapts its depth based on feature complexity, ensuring atomic tasks, traceable requirements, and mandatory verification checks. Use it for developing, validating, or improving features against precise specifications.

This skill provides advanced guidelines for designing robust and effective tools specifically for autonomous AI agents. It shifts the focus from traditional developer APIs to contracts optimized for Language Models (LLMs). Key topics covered include treating tool descriptions as prompt engineering, implementing namespacing, applying the Consolidation Principle, and utilizing Architectural Reduction patterns (like file system access) to maximize agent reliability and workflow complexity.

torchforge is Meta's advanced PyTorch-native library designed for agentic Reinforcement Learning (RL). It fundamentally separates core RL algorithms from complex distributed infrastructure concerns. This allows researchers to rapidly prototype and experiment with novel RL methods (like GRPO) using clean, PyTorch-native abstractions, while automatically handling large-scale tasks like distributed training across thousands of GPUs using Monarch and TorchTitan.