Skill UI

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.

This skill provides a comprehensive guide to understanding OpenRouter's API billing structure. It teaches users how to accurately calculate costs based on prompt (input) and completion (output) tokens. Users can utilize provided scripts to estimate costs, check credit balances, and implement cost-saving strategies like using `:floor` or `:free` model variants. Ideal for budgeting, performance comparison, and production cost tracking.

This skill implements a sophisticated, config-driven rules engine for OpenRouter API calls. It dynamically determines the optimal large language model (LLM) by evaluating a comprehensive context, including user subscription tier, remaining cost budget, specific task type (code, chat), required tools, and latency requirements. This is essential for building production-grade, cost-aware, and robust LLM orchestration systems.

This comprehensive guide provides developers with step-by-step instructions for migrating applications from direct calls to proprietary Large Language Model (LLM) APIs (such as OpenAI or Anthropic) to utilizing the unified OpenRouter gateway. It details essential code changes, including updating base URLs, managing API keys, adding custom headers, and correctly prefixing model identifiers. This ensures a seamless transition, enabling multi-model compatibility and simplified API management.

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.

Automates penetration test planning tasks by guiding workflows, validating outputs, and generating production-ready artifacts aligned with compliance and threat-modeling needs.

This guide provides advanced strategies for developers to significantly reduce Perplexity Sonar API billing costs. It covers key techniques including model routing (selecting the right model by query type), setting maximum output tokens, implementing robust query caching, and utilizing domain filters to minimize per-request charges. Essential for building cost-effective, high-volume AI applications.

This guide details how to implement Role-Based Access Control (RBAC) for Perplexity API in enterprise environments. Since Perplexity lacks native RBAC, the solution involves building a middleware gateway. This gateway enforces granular policies, including team-specific API key usage, model restriction (e.g., limiting access to 'sonar' only), budget caps, rate limiting, and mandatory search domain filtering for compliance and security.

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 provides advanced architectural patterns to optimize interactions with the Perplexity Sonar API. By implementing smart model routing, query hashing with time-to-live (TTL) caching, streaming responses, and rate-limited parallel processing, developers can significantly reduce latency, improve overall throughput, and manage API costs. Ideal for integrating complex, research-intensive search-augmented applications.

This skill provides comprehensive patterns for building highly reliable applications that interact with external Large Language Model (LLM) APIs, such as Perplexity Sonar. It implements critical reliability techniques including Circuit Breakers, Model Fallback, Streaming Timeout handling, and Caching layers. Use this to ensure your application maintains high availability and graceful degradation even when external services experience intermittent failures or variable performance.