7.1K skills · sorted by relevance
Build and check statistical models in PyMC while quantifying uncertainty. Supports hierarchical models that represent grouped data, sampling diagnostics, predictive checks, and model comparison, with templates and scripts for predictions and diagnostic reports.
Assess language models with over 60 academic benchmark tests using consistent prompts and scoring. Compare models, track performance during training, and organize research results with support for HuggingFace models, vLLM, and APIs.
Evaluate code generation models with HumanEval, MBPP, and multilingual benchmarks. The skill guides test execution and model comparisons using pass@k, an estimate of whether at least one of k generated solutions passes the tests.
Replace the object detector in VSS Alerts Blueprint verification mode with an ONNX model. This guide covers model placement, interpreting detection outputs, DeepStream configuration, TensorRT engine builds, redeployment, and common loading or mount failures.
Create and modify NeMo AutoModel training and evaluation recipes. Get guidance on YAML configuration, functions that build training components, execution flow, validation, and checkpoint saving and resuming, with minimal examples and local checks.
Integrate a machine learning weather model into Earth2Studio for forecasts that advance through time. The skill guides wrapper implementation, coordinate handling, model loading, tests, dependency declarations, and documentation updates using a reference inference script.
Build Earth2Studio wrappers for transformations at a single time point, from calculating wind speed to using pretrained or diffusion models. The workflow covers input and output coordinates, model loading, tests, and supporting documentation.
Integrate a supported object detection model from HuggingFace or NVIDIA NGC into NVIDIA DeepStream. Automates model acquisition and conversion, TensorRT engine building, detection output parsing, multi-stream performance benchmarks, and reports with charts in Markdown, HTML, and PDF.
Builds generative AI applications on Amazon Bedrock. Covers model invocation (Converse API, InvokeModel), RAG with Knowledge Bases, Bedrock Agents, Guardrails, and AgentCore (including the Harness managed agent loop). Applies when invoking models, setting up Knowledge Bases, creating agents, applying guardrails, deploying to AgentCore, migrating/porting/converting a Bedrock Agent (including inline agents) to an AgentCore Harness, troubleshooting Bedrock errors (ThrottlingException, AccessDeniedException), or choosing models (Claude, Llama, Nova, Titan). Also for prompt caching, quota and throttling diagnosis, cost tracking, migrating between Claude model generations (4.5 to 4.6 to 4.7), chunking strategies, API selection (Converse vs InvokeModel), guardrail capabilities, and model selection. Also covers AgentCore Payments (x402, microtransactions, Payment Manager, Connector, Instrument, Coinbase CDP, Stripe Privy, paid endpoints, agent payments). NOT for custom model training, Rekognition, or Comprehend.
The English summary is being prepared. Showing the original description.
Research, compare, and update shared AI model JSON for TypeScript, web, and Rust consumers. Covers text model tiers, image and video generation models, image tool models, release provenance, pricing data sourcing, and product-side catalogue cost estimates. Use when bumping model versions, adding new models, updating pricing, or auditing model specs against provider documentation.
The English summary is being prepared. Showing the original description.
Use the skill to control and verify the code interface configuration of your model — how Simulink® model elements are represented in the generated C or C++ code. This includes — (1) specifying names, types (storage classes), and placement of variables that represent model elements (for example, making a model parameter tunable as a global extern variable); (2) specifying names, types, and placement of functions that represent model algorithms; (3) selecting the deployment type (Component, Subcomponent, or Automatic); (4) selecting the interface configuration type (data or service interface); (5) linking a shared Embedded Coder dictionary to a model; (6) creating Embedded Coder dictionary entries and setting their properties; (7) for service interface configuration — specifying service interface definitions, including sender, receiver, client, and server services. Items 1 and 2 can be set per element or as a category-wide default. Item 1 applies to GRT and ERT models; the rest to ERT models only.
The English summary is being prepared. Showing the original description.
Generate C/C++ or CUDA code from an AI model (PyTorch, LiteRT) using MATLAB Coder or GPU Coder. Use when the user wants to integrate an AI model into an application with code generation as the end goal — generating MEX, CUDA MEX, static library, dynamic library, or executable — or using the model in Simulink for simulation and code generation. Covers PyTorch ExportedProgram (.pt2) via loadPyTorchExportedProgram and LiteRT (.tflite) via loadLiteRTModel (R2026a+). Keywords: PyTorch, torch, .pt2, ExportedProgram, loadPyTorchExportedProgram, invoke, codegen, MEX, CUDA, GPU, C, C++, deploy, AI model, deep learning model, LiteRT, TFLite, TensorFlow Lite, Simulink, slbuild, PyTorch ExportedProgram block, MATLAB Function block, dlosslib, loadLiteRTModel.
The English summary is being prepared. Showing the original description.
Use this skill for WeChat Mini Program AI via wx.cloud.extend.AI (小程序, wx.cloud apps). Covers generateText and streamText with callbacks (onText, onEvent, onFinish); streamText needs a data wrapper, generateText returns the raw response. Models via wx.cloud.extend.AI.createModel with groups hunyuan-exp (小程序成长计划), cloudbase (main managed), or custom-*; model id goes in the data wrapper `model` field. MUST run two-step preflight before code — see body. NOT for browser/Web (use ai-model-web), Node.js backend (use ai-model-nodejs), or image generation (use ai-model-nodejs).
The English summary is being prepared. Showing the original description.
Use this skill when a browser/Web app (React, Vue, Next, Nuxt, static sites, SPAs, dashboards, AI chat UI, 页面, 前端, 网页) needs AI models via @cloudbase/js-sdk. Default routing for Web/frontend AI — call directly from the browser, do NOT propose a Node.js proxy. Covers generateText and streamText; models via ai.createModel with groups cloudbase, hunyuan-exp, or custom-*, model id in the `model` field. MUST run two-step preflight before code — see body. NOT for Node.js backend (use ai-model-nodejs), Mini Program (use ai-model-wechat), or image generation (Node SDK only).
The English summary is being prepared. Showing the original description.
This skill should be used whenever the user mentions a "semantic model", "data model", or "dataset", or asks to "build", "model", "design", "optimize", "review", or "audit" one, or to "add a measure", "add a relationship", "create a role" / "set up RLS", "add a calculation group", "set up incremental refresh", "fix a star schema", "reduce model size", "prepare a model for Copilot / AI", or "check model quality". Covers the full lifecycle (design, build, refresh, review) and drives every operation through the `te` CLI first, then TOM (connect-pbid) or a model MCP, then TMDL authoring (the tmdl skill). Not for report visuals (use pbir-cli) or isolated DAX query tuning (use the dax skill).
The English summary is being prepared. Showing the original description.
Split a multi-call LM workflow by cognitive load, not by accuracy: let one strong model make the few reasoning decisions and a cheap model do the many mechanical executions (Aider architect+editor, DSPy optimizer-LM vs task-LM, vLLM speculative draft+target, LangGraph supervisor+worker are the same shape). Use when designing or cost-optimizing a pipeline that calls an LM many times, when deciding which steps need a strong reasoner vs a cheap executor, or when adding an escalation valve for when the cheap tier degrades. Search keywords: reduce LLM cost, cheaper model, lower token cost, model cascade, route to cheap model, strong model plus cheap model, LLM cost optimization.
The English summary is being prepared. Showing the original description.
Lifecycle SOP for **per-model prompt artifacts** — the compiled prompts, instructions, few-shot demos, edit-format pins, and embedding-bound indices that change behavior when the underlying LM, dataset, or framework version changes. Activate when adopting compiled prompts (DSPy, GEPA, BootstrapFewShot output), when supporting multiple LMs in production, when a provider deprecates a model snapshot, or when a framework deprecates a config surface (LlamaIndex `ServiceContext` → `Settings`, Aider edit-format defaults). Do NOT activate for one-off raw prompt edits or for truly model-agnostic system prompts that have been swap-tested. Search keywords: prompt portability, model swap, recompile prompt, model deprecation, prompt per model, prompt breaks on new model, version compiled prompts.
The English summary is being prepared. Showing the original description.
Generate a coherent, whole-paper mathematical modeling solution framework from a complete contest problem. Use after the user supplies a CUMCM or other modeling problem and asks for modeling ideas, an overall solution plan, question-by-question analysis, candidate-model comparison, model-selection reasons, innovations, validation, or cross-question linkage. For every question, explain the task and mathematical essence, compare multiple feasible models in tables, recommend a route with explicit reasons, and keep all questions connected through shared data, variables, parameters, constraints, and validation.
The English summary is being prepared. Showing the original description.
Register trained models in MLflow Model Registry with version control, implement stage transitions (Staging, Production, Archived) with approval workflows, and manage model lineage with comprehensive metadata and deployment tracking. Use when promoting a trained model from experimentation to production, managing multiple model versions across development stages, implementing approval workflows for governance, rolling back to previous versions, or auditing model changes for compliance.
The English summary is being prepared. Showing the original description.
Bridge Codex to web-based AI model products by packaging local task context, scrub-checking it, sending it through an approved browser surface to ChatGPT Pro, Claude, Grok, Gemini, or another web model, waiting for the answer, and returning the model response to the user or Codex. Also guide DevSpace-like MCP Connector Mode when the user wants ChatGPT Pro or another MCP-capable web host to access approved local workspaces without relying on Codex browser automation. Use when the user asks for GPT Pro, ChatGPT Pro, Claude web, Grok, Gemini web, web model bridge, external model consult, ask another model, second opinion, use a web AI model, browserless agent access to GPT Pro, MCP connector, DevSpace-like workflow, or send local repo context to a browser-based model for planning, review, debugging, architecture discussion, or implementation guidance.
The English summary is being prepared. Showing the original description.
Guide for choosing models in Tinker — available model families, model types (base, instruction, reasoning, hybrid, vision), architecture (dense vs MoE), and how to match renderers to models. Use when the user asks which model to use, what models are available, or how to pick a model for their task.
The English summary is being prepared. Showing the original description.
Investigate model specifications, requirements, and evaluation strategy. Use when researching a model before benchmarking: reading HuggingFace model cards, estimating VRAM requirements, selecting GPU providers, and determining evaluation approach. Triggers on "model research", "investigate model", "model info", "VRAM estimate", "which provider", "model card".
The English summary is being prepared. Showing the original description.
Model interpretability and explainability using SHAP (SHapley Additive exPlanations). Use this skill when explaining machine learning model predictions, computing feature importance, generating SHAP plots (waterfall, beeswarm, bar, scatter, force, heatmap), debugging models, analyzing model bias or fairness, comparing models, or implementing explainable AI. Works with tree-based models (XGBoost, LightGBM, Random Forest), deep learning (TensorFlow, PyTorch), linear models, and any black-box model.
The English summary is being prepared. Showing the original description.
Find the cheapest AI model that is smart enough for a task, and the cheapest provider to run it on, from live Artificial Analysis benchmarks via Firecrawl. Use when the user asks which model to use, how to cut their LLM or API bill, whether a cheaper model would do, which provider is cheapest or fastest for a model, which open-weights model to self-host, or wants something built with the cheapest model for each step.
The English summary is being prepared. Showing the original description.