Development skills

Browse reusable Agent Skills, each with a clear purpose and practical guidance.

add-diffusion-model

Add a new diffusion model (text-to-image, text-to-video, image-to-video, text-to-audio, image editing) to vLLM-Omni, including Cache-DiT acceleration and parallelism support (TP, SP/USP, CFG-Parallel, HSDP). Use when integrating a new diffusion model, porting a diffusers pipeline or a custom model repo to vllm-omni, creating a new DiT transformer adapter, adding diffusion model support, or enabling multi-GPU parallelism and cache acceleration for an existing model.

5.59k repo starsObserved in 1 repos

Development

GitHub

add-tts-model

Integrate a new text-to-speech model into vLLM-Omni from HuggingFace reference implementation through production-ready serving with streaming and CUDA graph acceleration. Use when adding a new TTS model, wiring stage separation for speech synthesis, enabling online voice generation serving, debugging TTS integration behavior, or building audio output pipelines.

5.59k repo starsObserved in 1 repos

Development

GitHub

quantization

Work on vLLM-Omni quantization for diffusion, autoregressive, omni, or multi-stage models. Use when choosing or adding methods such as fp8, int8, gguf, mxfp8, mxfp4, mxfp4_dualscale, ModelOpt, AutoRound, INC, msModelSlim, awq, or gptq; debugging quantized loading; or validating memory, speed, and output quality.

5.59k repo starsObserved in 1 repos

Development

GitHub

add-to-project

Add addons or features (PWA, Tauri, Starlight/Fumadocs docs, Biome/Oxlint, Husky/Lefthook, Turborepo/Nx, the MCP addon, etc.) to an existing Better-T-Stack project. Use when the user wants to extend, enhance, or add tooling to a project that was created with Better-T-Stack.

5.58k repo starsObserved in 1 repos

Development

GitHub

scaffold-project

Scaffold a new app, API, backend, fullstack project, monorepo, or starter with Better-T-Stack — including new projects built on a specific framework like Hono, Express, Fastify, Elysia, Next.js, TanStack Router/Start, Nuxt, Svelte, Solid, Astro, or React Native (native-bare, native-uniwind, native-unistyles). Use whenever the user wants to start, create, bootstrap, set up, or initialize a new project/app/API, e.g. "create a Hono app", "start a Next.js project", "scaffold a fullstack app with auth and a database". Prefer generating the project through the Better-T-Stack MCP server (plan then create) over hand-writing package.json, config, and folders.

5.58k repo starsObserved in 1 repos

Development

GitHub

azure-ai-projects-emit-from-typespec

Emit the azure-ai-projects Python SDK from TypeSpec, apply post-emitter fixes, and create a Pull Request. WHEN: "emit SDK from TypeSpec", "generate azure-ai-projects SDK", "update azure-ai-projects from TypeSpec", "emit from TypeSpec", "regenerate azure-ai-projects". DO NOT USE FOR: other Azure SDK packages, manual code edits without TypeSpec. INVOKES: PostEmitter.ps1 script, git commands, gh CLI for PR creation.