Loading...
Loading...
Found 111 Skills
Skill for full-cycle post-loan risk management of credit customers. Execute post-loan inspection, fund usage verification, financial health assessment, guarantee validity review, early warning identification and classification adjustment, and closed-loop early warning disposal. Trigger words include: "post-loan inspection", "post-loan management", "risk classification adjustment", "post-loan inspection report", "conduct a post-loan inspection", "check post-loan management", "does this customer's risk classification need adjustment?", "early warning disposal plan", "fund usage verification". Not applicable to: pre-loan due diligence, credit approval, post-loan management of personal credit, pure financial analysis not involving post-loan scenarios.
Connects to and performs inference with Google Cloud Agent Platform GenAI models, including First-Party Gemini models and Third-Party OpenMaaS models (Llama, DeepSeek, Qwen, etc.). Use when you need to generate code for calling Gemini or OpenMaaS models, authenticate with GenAI SDK, OpenAI SDK, or legacy Agent Platform SDK, configure base URLs and global/regional endpoints, or troubleshoot 429 Resource Exhausted (DSQ), 400 User Validation, or 404 Not Found errors. Don't use for deploying models to endpoints or for running model evaluations.
LangChain / LangGraph engineering pitfalls and verified fixes. Covers DeepAgents, OpenAI-compatible model integration (including Chinese provider adapters: DeepSeek, Qwen, GLM, etc.), middleware, streaming, multi-agent orchestration, and other common development issues. Use when hitting unexpected behavior, making architecture decisions, or integrating Chinese LLM providers during LangChain development.
Implements the NOWAIT technique for efficient reasoning in R1-style LLMs. Use when optimizing inference of reasoning models (QwQ, DeepSeek-R1, Phi4-Reasoning, Qwen3, Kimi-VL, QvQ), reducing chain-of-thought token usage by 27-51% while preserving accuracy. Triggers on "optimize reasoning", "reduce thinking tokens", "efficient inference", "suppress reflection tokens", or when working with verbose CoT outputs.
CLI tools execution specification (gemini/claude/codex/qwen/opencode) with unified prompt template, mode options, and auto-invoke triggers for code analysis and implementation tasks. Supports configurable CLI endpoints for analysis, write, and review modes.
BYOK — register a custom LLM endpoint (Anthropic, OpenAI, Qwen, DeepSeek, etc.) with your own API key
PixVerse CLI — generate AI videos and images from the command line. Supports PixVerse, Veo, Sora, Kling, Hailuo, Wan, and more video models; Nano Banana (Gemini), Seedream, Qwen image models; and PixVerse's rich effect template library. Start here.
Novita AI: LLM, Image Generation & Editing, Video Generation, Audio (TTS/ASR), and GPU Cloud. Use this skill whenever the user wants to call Novita AI APIs — chat with LLMs (DeepSeek, Llama, Qwen), generate images (FLUX, Stable Diffusion, Seedream, Hunyuan Image), edit images (remove background, upscale, inpainting, img2img, outpainting, reimagine, merge face, replace background, remove text), generate videos (Kling, Wan, Hunyuan, Minimax Hailuo, Vidu, PixVerse, Seedance), do text-to-speech or speech-to-text (MiniMax TTS, GLM TTS, Fish Audio, ASR, voice cloning), run OpenAI-compatible batch jobs, manage GPU cloud instances and serverless endpoints, or check account balance and billing. Also trigger when the user mentions novita.ai, Novita AI, Novita API key, or wants to use any Novita platform service — even if they just say "generate an image" or "run an LLM" and Novita is available as a provider.
Choose the right MoE token dispatcher (`alltoall`, DeepEP, or HybridEP) for the hardware, EP degree, and optimization stage. Summarizes patterns from DSV3, Qwen3, Qwen3-Next, and VLM bring-up work.
Write, review, and improve prompts for any LLM — Claude, GPT, Gemini, Llama, DeepSeek, Mistral, Cohere, Qwen, Grok, Nova, and more. Use when the user asks to "write a system prompt", "improve this prompt", "review my prompt", "make a prompt for", "optimize my prompt", "fix my prompt", "why isn't my prompt working", or wants help writing better prompts for any AI model. Also use when building agents, chatbots, or AI assistants that need system-level instructions, or when the user has a bad prompt they want rewritten. Covers system prompts, task prompts, tool descriptions, and general prompt improvement across all major model families.
Build and publish a Gradio demo on Hugging Face Spaces for a user-provided LoRA. Use when someone asks to create, generate, ship, or publish a Space, demo, Gradio app, or playground for a LoRA — including LoRAs for Qwen-Image, Qwen-Image-Edit, LTX-Video, Wan, FLUX, SDXL, or other diffusion base models. Also triggers when someone describes a LoRA they trained or hosts on the Hub and wants to share it. Covers picking the right base pipeline and `diffusers` inference recipe, designing a UI tailored to the LoRA's task and inputs (Union/multi-task control, edit, video, image, etc.), respecting model-card recommendations (trigger words, steps, guidance, LoRA scale, example inputs), and shipping to ZeroGPU hardware as a private Space by default.
Provides guidance for writing and benchmarking optimized CUDA kernels for NVIDIA GPUs (H100, A100, T4) targeting HuggingFace diffusers and transformers libraries. Supports models like LTX-Video, Stable Diffusion, LLaMA, Mistral, and Qwen. Includes integration with HuggingFace Kernels Hub (get_kernel) for loading pre-compiled kernels. Includes benchmarking scripts to compare kernel performance against baseline implementations.