Skill UI

This guide documents common mistakes, gotchas, and anti-patterns encountered during Kling AI API integration. It covers critical best practices regarding data types, token management, polling, URL persistence, and prompt engineering to ensure stable and reliable AI video generation workflows. Essential reading for robust development and troubleshooting.

This comprehensive guide provides strategies for optimizing video generation using Kling AI. It covers the crucial trade-offs between speed, quality, and cost by selecting the optimal model and parameters. Techniques detailed include performance benchmarking, implementing connection pooling, advanced caching strategies, and prompt engineering best practices to ensure efficient and high-quality outputs for any production use case.

This guide outlines common anti-patterns and critical mistakes when using the Windsurf AI IDE. It covers best practices for prompt engineering, managing workspace scope (especially in monorepos), distinguishing between Supercomplete and Cascade usage, and ensuring safe, maintainable AI-assisted coding workflows. Essential for new developers and team auditors.

This comprehensive guide details the credit-based usage model of the Windsurf AI platform. It explains how various powerful models (e.g., SWE-1, GPT-4o) consume credits and provides strategies for optimizing AI usage costs. Learn how to monitor remaining credits, understand different pricing tiers (Free, Pro, Teams), and implement prompt engineering techniques to maximize efficiency and prevent mid-session interruptions.

A comprehensive guide to establishing production-ready coding standards and context for the Windsurf AI environment. Learn how to utilize root-level rules (.windsurfrules), granular workspace rules, Model Context Protocol (MCP) server integration, and advanced prompt engineering to guide the AI and maintain consistent code quality across projects.

Security guardrails embedded in the mortgage plugin that block prompt injection and social-engineering tricks, stop system prompt extraction, guard PII and pricing logic, enforce workflow phases, and ensure conversational inputs stay within refinance scope.

A comprehensive collection of six expert-level, structured workflows designed to elevate AI assistance from ad-hoc querying to repeatable, production-grade engineering practices. Users can evaluate prompts, design RAG architectures, plan context budgets, run security audits, build eval harnesses, and coach product strategy.

An expert guide for reducing large-scale LLM API costs (40-80%). Learn engineering techniques like model routing, prompt caching, and output length control to make every token count. Essential for engineers building cost-efficient AI features and robust endpoints.

A specialized prompt engineering assistant that optimizes and generates production-ready prompts for various AI tools, including LLMs, Midjourney, and coding agents. It enforces strict rules, extracts detailed intent (Task, Context, Audience, etc.), and outputs optimized, single-block prompts, ensuring zero wasted tokens and professional quality output.

This skill provides a guided, hands-on tutorial to teach users the Ouroboros workflow. It guides the user through a structured process of refining vague ideas into precise, executable specifications (Seeds). By simulating an interview and exposing hidden assumptions, it reduces the inherent ambiguity in initial project planning, making it ideal for product development, AI prompt engineering, and initial concept validation.