Skill UI

This skill guides developers on migrating their codebase from direct OpenAI API calls to using OpenRouter. By making minimal changes—primarily updating the base URL and API key—users gain access to over 400 models from multiple providers (including Anthropic and Google) through a unified SDK interface, ensuring high compatibility and flexible model switching.

This skill provides comprehensive best practices and Python code examples for optimizing API calls to Large Language Models (LLMs) via OpenRouter. It covers benchmarking model latency, implementing streaming to achieve low Time-To-First-Token (TTFT), and processing requests concurrently using asynchronous programming (asyncio). Use this when developing real-time, high-throughput applications that require predictable and fast model responses.

Flash Attention optimizer delivers 2-4x speedup and 10-20x memory savings for transformer attention, ideal when training or running long-context models that hit GPU memory or latency limits; supports PyTorch native scaled_dot_product_attention, the flash-attn library, H100 FP8 precision, and sliding-window attention.

Optimizes deep learning models by analyzing architecture, data, and metrics to select optimizers, learning rate schedules, and regularization strategies, then outputs optimized code and evaluates accuracy, time, and resource usage.

Enables regression modeling via the regression-analysis-tool plugin by analyzing datasets, choosing linear or polynomial fits, validating models, and reporting metrics so users can predict outcomes or explore variable relationships.

This comprehensive runbook provides step-by-step procedures for rapidly responding to Perplexity API outages and service degradations. It covers initial triage (checking models, auth, DNS), handling common errors (401, 429, 500/503), implementing fallbacks, and structured post-incident activities, including evidence collection and postmortem documentation.

This skill provides a comprehensive guide to configuring the Perplexity Sonar API across distinct operational environments (development, staging, and production). It details how to manage environment-specific API keys, enforce tailored rate limits, select appropriate models (e.g., sonar vs. sonar-pro), and implement cost optimization strategies. Use this when migrating or integrating Perplexity search functionality into a multi-tier, production-grade application architecture.

This guide implements comprehensive guardrails for Perplexity Sonar API. It covers crucial aspects like query content moderation, PII sanitization, enforcing user usage quotas, selecting appropriate models based on user tiers, and scoring citation reliability to ensure safe, ethical, and cost-effective AI usage.

This comprehensive guide assists developers in migrating their code and API calls when Perplexity updates its model generations or changes API parameters. It provides a clear map from legacy models (pplx-api) to the current Sonar family (sonar, sonar-pro), covering model name changes, base URL updates, parameter adjustments, and SDK upgrade steps. Use this guide whenever you encounter deprecation warnings or need to ensure compatibility with the latest Perplexity Sonar models.

A guide to building robust projections and read models from continuous event streams. This pattern is essential for implementing CQRS read sides, creating materialized views, optimizing query performance, and building complex search indexes in event-sourced architectures. Use it when data needs to be transformed and aggregated from raw events into optimized query structures.

This skill specializes in designing, optimizing, and evaluating prompts for Large Language Models (LLMs). It allows users to generate optimized prompt templates, structured output schemas (e.g., JSON mode, function calling), and comprehensive test suites. Use this skill when building new LLM applications, implementing complex reasoning patterns like Chain-of-Thought, defining system guardrails, or building robust evaluation frameworks to measure and improve model performance.

A comprehensive collection of battle-tested, structured prompts designed to elevate interactions with Large Language Models (LLMs). Covers advanced use cases including expert coding reviews, system architecture design, performance analysis, technical writing, and creative brainstorming, ensuring high-quality, actionable AI output.