Skill UI

This is an all-in-one command-line tool for generating and editing images using the full Nano Banana model family. Users can select the optimal model—Gemini 3.1 Flash for balanced performance, Gemini 3 Pro for high-quality output, or Gemini 2.5 Flash for speed—to suit their needs. Key features include text-to-image generation, precise single image editing, and multi-image collage composition, making it versatile for creative and technical workflows.

This skill provides advanced architectural patterns for optimizing high-volume integrations with the Navan API. It addresses common bottlenecks like rate limiting, sequential calls, and naive pagination by implementing robust solutions such as data-type-specific response caching, controlled parallel fetching with concurrency throttling, efficient cursor-based pagination, and HTTP connection reuse. Use this when building scalable, reliable, and high-throughput systems.

This guide provides advanced strategies for optimizing interactions with the Notion API. Learn best practices such as implementing Time-To-Live (TTL) caching, batching multiple block updates, avoiding inefficient N+1 query patterns, and managing rate limits. Use this when your Notion application suffers from slow responses, high latency, or excessive API call consumption.

A comprehensive guide for optimizing OneNote Graph API interactions at scale. It addresses common performance bottlenecks such as slow data retrieval from large notebooks, handling excessive image sizes, implementing efficient payload selection ($select), and paginating massive data sets using $top. Essential for building stable, high-performance automation workflows.

This skill provides a deep dive into optimizing Oracle Cloud Infrastructure (OCI) resources. It guides users through selecting appropriate compute shapes (e.g., E5.Flex, A1.Flex) and understanding the performance tiers for block volumes (Balanced, Ultra High Performance). Users can programmatically query available shapes, create performance-tuned volumes, and monitor real-time IOPS and network bandwidth using the OCI SDK, ensuring optimal infrastructure decisions for any workload.

Qdrant is a high-performance vector database. This client SDK suite provides official, easy-to-use libraries for multiple programming languages (including Python, Java, Go, and TypeScript). It enables developers to seamlessly integrate sophisticated vector search capabilities into their applications, supporting both REST and gRPC APIs. Ideal for building semantic search engines, recommendation systems, and advanced AI applications.

This guide provides comprehensive guidance on scaling Qdrant, a vector database. It helps users diagnose performance bottlenecks and determine the optimal horizontal or vertical scaling strategy. Strategies are categorized based on specific needs: increasing data volume, improving query throughput (QPS), minimizing query latency, or managing high query result volumes.

This comprehensive guide provides advanced strategies for optimizing and scaling Qdrant vector database throughput (QPS). It covers performance tuning techniques, such as optimizing segment size, enabling quantization, and using batch search. Furthermore, it details horizontal scaling methods using read replicas and addressing disk I/O bottlenecks, ensuring high performance for large-scale, high-concurrency retrieval applications.

This mechanism is designed to optimize Qdrant's performance when handling queries with very large limits or when paginating results across multiple shards. Instead of forcing every shard to transfer the full requested limit (which is bandwidth-intensive), it uses Poisson distribution statistics to calculate smaller, optimized limits for each shard. This significantly reduces inter-shard data transfer and improves the efficiency of large-scale vector search, while maintaining a very high degree of accuracy.

A comprehensive guide for diagnosing and resolving slow search performance in Qdrant. This resource covers issues related to low QPS (throughput), high latency for single queries, and slow filtered searches. Solutions include tuning HNSW parameters, managing memory pressure, implementing quantization, optimizing indexing, and configuring cluster settings for maximum efficiency.

This guide details performance tuning and implementation patterns for the RunwayML platform. It provides a practical Python SDK pattern for advanced AI tasks, including image-to-video conversion and generating high-quality creative content using various models like gen3a_turbo. Ideal for developers integrating creative AI capabilities into their applications.

This guide provides comprehensive methods for load testing and capacity planning specifically for Salesforce integrations. It walks users through calculating API limit budgets, running high-volume performance tests using tools like k6, and optimizing Bulk API throughput. Use this when scaling complex, high-volume, or mission-critical Salesforce processes.