Total 53,884 skills, AI & Machine Learning has 8967 skills
Showing 12 of 8967 skills
Expert skill for using MOSS-TTS-Nano, a 0.1B parameter multilingual real-time TTS model that runs on CPU with voice cloning support.
Feed-forward 3D foundation model for streaming scene reconstruction using Geometric Context Transformer
Stability AI integration. Manage data, records, and automate workflows. Use when the user wants to interact with Stability AI data.
Architecture patterns and best practices for giving AI agents email capabilities. Use when designing how agents send, receive, and manage email conversations, building two-way communication loops, implementing human-in-the-loop approval with drafts, choosing between WebSockets and webhooks, setting up multi-agent email topologies, handling OTP and verification flows, or securing agent email against prompt injection.
Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.
AI-powered web search via Exa with content extraction. Use when user says "exa search", "web search with content", "find similar pages", or needs broad web results beyond academic databases (arXiv, Semantic Scholar).
Export agent data into a Starchild migration bundle. For use by ANY agent (OpenClaw, Claude Code, Cursor, etc.) to migrate into Starchild.
Code of Conduct for De-Buzzwording Chinese Output. Automatically activated when the model responds in Chinese to ensure concise, direct, and natural output, avoiding AI-style expressions such as consulting buzzwords, internet slang, false sense of urgency, and emotional manipulation. Trigger words: 'Speak human language', 'Cut the fluff', 'Remove buzzwords', 'Get rid of AI tone', 'Don’t be so GPT-like', 'Speak normally'. Even without trigger words, the rules of this Skill shall take effect whenever the model generates Chinese responses. It also applies to scenarios where users say 'This text is too GPT-like', 'Help me rewrite it in human language', 'The output is too cheesy', 'Stop beating around the bush', etc.
Expert skill for using TileKernels, a library of optimized GPU kernels for LLM operations (MoE routing, quantization, transpose, engram gating, Manifold HyperConnection) built with TileLang.
System prompt toolkit that removes AI slop and makes any LLM respond like a normal person — concise, direct, no filler.
Annie Duke's Decision Quality framework applied to a business decision. Spawns a team of specialist agents — Resulting Auditor, Calibrator, Pre-Mortem Analyst, Quit Strategist, Process Architect — who each apply a distinct lens from Duke's framework to evaluate whether a decision is sound regardless of outcome. The lead synthesizes into a stacking analysis: which biases are operating, which process flaws exist, and the honest Duke verdict. Use when the user says "duke this", "is this a good bet", "should I quit", "evaluate this decision", or faces any high-stakes choice under uncertainty and wants rigorous decision-process analysis. Works as a standalone analysis or after /office-hours.
Trigger native web search. Use when you need quick internet research with concise summaries and full source URLs.