Skill UI

This skill demonstrates advanced techniques for managing high-volume API calls to LLM services via OpenRouter. It implements multi-key round-robin strategies, incorporating health checks and circuit breakers to distribute load across multiple API keys. It also covers concurrent request processing using asyncio and provider-level failover, ensuring high availability and maximizing throughput when facing rate limits.

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.

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.

Automate Composio’s PDF API IO workflows through the Rube MCP connector by discovering current tool schemas, managing the pdf_api_io connection, and executing RUBE_MULTI_EXECUTE_TOOL calls with schema-compliant arguments for reliable PDF operations.

A comprehensive guide to identifying and avoiding common integration mistakes and anti-patterns when using the Perplexity Sonar API. It covers critical best practices such as correctly handling citations, setting maximum token limits, utilizing recency filters, and optimizing API calls to ensure cost efficiency and data accuracy. Essential for code review and developer onboarding.

A comprehensive guide and implementation examples for handling API rate limits (HTTP 429) when interacting with the Perplexity Sonar API. This resource provides multiple robust techniques, including exponential backoff with jitter, queue-based rate limiting, and custom Token Bucket implementation, ensuring reliable and high-throughput API calls for developers using TypeScript and Python.

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.

Provides a comprehensive workflow for managing feature flags, running A/B/multivariate experiments, and analyzing user cohorts using PostHog. The guide covers implementing flag evaluation both client-side (browser) and server-side (Node.js), and demonstrates how to programmatically set up, track, and query experiment results via API calls (cURL/Bash).

This guide provides minimal working examples demonstrating the three core PostHog functionalities: event capture, user identification, and feature flag evaluation. It covers both server-side (Node.js, Python) and client-side (JavaScript) SDK usage, as well as direct API calls, making it perfect for initial integration setup, testing, and learning basic product analytics patterns.

A workflow for orchestrating Pushover operations through Composio’s Rube MCP connectors, covering tool discovery, connection validation, schema-aware execution, and recommended API calls for reliable automation.