Performance skills

This skill provides a complete reference for building Miro Web SDK plugins and desktop apps, covering core APIs, authentication, events, and best practices.

This skill runs a 29-point GEO site readiness audit to assess AI accessibility, structured data, content citability, and technical setup for your website.

This skill enables unified access to 343+ models for text, image, JSON, video and no-deploy workflows, maximizing token efficiency and speed.

This skill provides fast TTS and STT with multiple voices and languages, enabling high-performance audio processing for Python apps.

This skill helps design an interruption-resilient AWS EC2 Spot strategy with diversified families and robust fallbacks for cost optimization.

This skill aggressively stress-tests LLM throughput by iterative prompt-tool-LMM cycles to maximize token consumption while monitoring usage.

This skill helps fetch LinkedIn data using the LinkdAPI Python SDK, enabling quick profile, company, and job insights via ephemeral uv scripts.

This skill optimizes macOS for AI workloads by tuning memory, UI, Docker, and SSH to turn an 8GB Mac into a near-16GB AI node.

This skill optimizes Feishu document editing by batching operations, preserving formatting, and handling large content with minimal API calls.

This skill processes EvoMap heartbeat work opportunities, applying formal verification, performance optimization, and concurrent systems techniques to deliver

This skill helps you resize, convert, and watermark images efficiently using libvips, reducing memory usage and speeding batch processing.

This skill reviews Q/kdb+ code for bugs, performance, and security, helping quant teams write correct, efficient kdb+ queries.

This skill intelligently routes users' requests to appropriate LLMs by complexity, saving costs by using cheaper models for simple tasks.

This skill helps you generate high quality speech, transcriptions, and voice conversions using EachLabs models for seamless audio workflows.

This skill helps you maximize token efficiency in Claude Max with a 10-step plan to extend weekly limits while preserving work output.

This skill generates a single high-performance X tweet for builders, delivering sharp, postable output with strict formatting and creator mode.

This skill audits OpenClaw token usage, trims workspace files, enables caching, and presets a cost-efficient memory and model strategy to reduce tokens.

This skill upscales videos to 720p, 1080p, 2k, or 4k using WaveSpeed AI, producing higher-resolution results from a public video URL.

This skill enables fast local transcription using faster-whisper with GPU acceleration, providing word-level timestamps and flexible model options for accurate

This skill enforces mandatory pre-deployment quality assurance by validating builds, testing browsers, ensuring mobile responsiveness, and verifying links and

This skill performs fast local speech-to-text transcription using Faster Whisper with NVIDIA GPU acceleration, delivering private, multi-language outputs.