Performance skills

This skill sends images as Feishu stickers by auto-uploading, caching, and optimizing for fast delivery and smart search.

This skill helps you compare crypto payment gateways and evaluate custody, KYC, fees, and sovereignty to choose the best solution.

This skill seamlessly selects and switches to the strongest available model in your plan with sub-100ms latency and zero restarts.

This skill enables OpenLang agent-to-agent communication to compress messages by 5-10x, reducing token usage and speeding inter-agent workflows.

This skill unifies searches across Google, Wikipedia, Reddit, and RSS with Redis caching to deliver fast, comprehensive results.

This skill deploys the Guild Stack with PostgreSQL indexing to enable sub-second game state queries for combat, scouting, and analytics.

This skill provides local speech-to-text using Parakeet or Whisper backends for accurate English or multilingual fast transcription.

This skill intelligently routes sub-agent tasks to the most suitable model based on complexity and cost to balance quality and savings.

This skill helps you drive HeartClaws strategy by automating API interactions, state reads, and action submissions for open world and quick matches.

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 routes each request to the cheapest capable model across 30+ providers to cut inference costs and speed.

This skill generates AI images using Google's Gemini 3.1 Flash via Evolink, enabling fast text-to-image and image editing with a single API key.

This skill automatically routes tasks to the optimal model by complexity, saving cost while preserving performance for sub-agent tasks.

This skill provides persistent, semantic memory for AI agents by storing and recalling important context across sessions with fast, local processing.

This skill enables autonomous self-improvement by safely updating configurations, skills, prompts, and memory to continuously optimize performance.

This skill reviews code for security, performance, and reliability across languages, delivering actionable findings to accelerate safe PR merges.

This skill helps you build fast, single-file Phaser 3 arcade games for mobile-first Remix/Farcade deployments with minimal boilerplate.

This skill guides a separate reviewer through a comprehensive HyperEVM dApp QA checklist, ensuring no common shipping mistakes are missed.

This skill provides world-class DJ technique guidance, beatmatching, harmonic mixing, transitions, and crowd-reading strategies to elevate performances.

This skill detects your local hardware and recommends the best-fit LLM models with optimal quantization and speed estimates.

This skill analyzes code quality across languages, delivering actionable improvements for readability, maintainability, performance, and security.