Skill UI

Covers prompt and response caching plus Cache Augmented Generation patterns for LLMs such as Claude and OpenAI to reuse stable context, reduce repeat work, and cut latency/cost on recurring queries.

Meta’s Prompt Guard is an 86M-parameter classifier that spots prompt injections and jailbreak attempts before they reach LLMs, filtering user input, third-party data, or RAG documents with multilingual support and sub-2ms latency for fast deployments.

Execute controlled chaos experiments to proactively test system resilience, fault tolerance, and recovery capabilities. This toolkit simulates real-world failures, such as network latency, service crashes, resource exhaustion, and dependency outages, ensuring that microservices degrade gracefully and recover automatically. It is essential for modern distributed systems validation.

This skill guides the setup of distributed tracing and performance monitoring using Sentry. It covers configuring the SDK (e.g., setting tracesSampleRate) to capture latency, identify performance bottlenecks, and trace requests across the entire application stack. It demonstrates how to create manual spans for business logic and utilizes profiling for deeper performance insights, ensuring comprehensive application observability.

A comprehensive guide covering observability patterns for service meshes like Istio and Linkerd. Learn how to implement the three pillars—Metrics, Traces, and Logs—to monitor service health, visualize dependencies, track golden signals (latency, traffic, errors, saturation), and debug microservice communication issues across complex distributed systems.

Serve production-grade LLM APIs with vLLM, leveraging PagedAttention, continuous batching, quantization, and tensor parallelism to maximize throughput while keeping latency, GPU usage, and scalability under control for OpenAI-compatible deployments.

This framework provides a comprehensive methodology for defining, measuring, and implementing Service Level Indicators (SLIs), Service Level Objectives (SLOs), and managing error budgets. It helps engineering teams quantify service reliability (e.g., availability, latency) and establish clear, measurable targets (SLOs). By tracking error budgets, teams can strategically balance rapid feature development (innovation velocity) against maintaining high system stability and reliability.

This skill provides comprehensive observability for the Speak language learning platform. It monitors critical metrics such as API health status, assessment latency, session usage, and detailed pronunciation score distributions. It is essential for developers implementing robust, production-grade monitoring systems and managing large-scale language learning operations.

This guide provides a comprehensive method for optimizing the latency and reliability of the Speak API integration. It covers techniques such as audio preprocessing, implementing response caching, and managing connection pooling. Use this skill when establishing a production-grade, low-latency language learning platform or managing high-volume API operations.

Speeds up LLM inference by combining speculative decoding with Medusa multiple heads and lookahead techniques so chatbots, code assistants, and other real-time workloads can run with 1.5-3.6× lower latency while preserving target-model quality.

This comprehensive guide provides systematic methods to optimize Supabase query and database performance. Techniques cover PostgreSQL indexing (basic, composite, partial), using EXPLAIN ANALYZE for query plan analysis, optimizing client calls (column selection, pagination), and managing infrastructure resources like connection pooling. Use this when facing slow queries, high latency, or preparing a project for production-scale traffic.

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.