Administer the per-prompt caveman/ponytail router — show status, force a mode, or turn routing off.
Deterministic mode selection for session-start. Reads Phase A STATE.md recommendations + (future) learnings, sessions, backlog, bootstrap signals and returns {mode, rationale,…
MoDeck integration. Manage data, records, and automate workflows. Use when the user wants to interact with MoDeck data.
Für Modehäuser: Dokumentenmatrix, Lückenliste und Nachforderung: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Schnittstellenkarte mit Zuständigkeits- und Nachweisfragen.
Use when you need to measure how much a specific model capability or architectural feature contributes to prediction performance, especially when that capability is non-obvious or…
Automatically appends the model alias to the end of every response with integrated hook functionality and configuration change detection.
Use when validating or evaluating a trained medical-imaging model. Audits split leakage and validation design, computes task-correct held-out metrics (Dice + HD95, AUROC + AUPRC,…
Use when the user asks to "audit the model", "check model health", "give me project stats", "what's in this model", "summarize the current view", "analyze the selection", or wants…
Monthly LLM stack audit — compare your current models against latest benchmarks and pricing from OpenRouter.
Run a 3-leg evaluation against a newly released LLM. Leg 1 is trap questions with deterministic answers, leg 2 is an agentic orchestration task with a planted failure, leg 3 is a…
Identifying, measuring, and mitigating algorithmic bias to ensure equitable outcomes in AI systems.
Builds and updates DCF, LBO, and 3-statement financial models in Excel with live data connections. Use when user says "build DCF", "create LBO model", "populate 3-statement model".
Refresh MODELS.md against current model releases, pricing, availability, and task-specific benchmark results.
Use when assessing how well a survival model's predicted probabilities agree with observed outcomes by fitting a Cox model and generating bootstrap calibration curves at one or…
Model card generation expertise — auto-generates Model Cards covering intended use, limitations, and training data provenance per the HuggingFace and Mitchell et al. standard
Choose the right model endpoint for a given task. Modality-organized catalog of production endpoint defaults, text-to-image, image-to-image, text-to-video, image-to-video, and…
Fetch CURRENT model information from live provider docs (Anthropic, OpenAI, Google Gemini, Ollama) and return a structured catalog the user can use to populate or validate the…
Model Checkpoint Manager - Auto-activating skill for ML Training. Triggers on: model checkpoint manager, model checkpoint manager Part of the ML Training skill category.
Compare AI models across benchmarks, pricing, and capabilities. Use when the user asks to compare models, evaluate trade-offs, or pick the best model for a task.
Use when when you have multiple candidate spectrum prediction models (e.g., FFN vs. GNN encoders, NEIMS vs. MassFormer vs.
Use when asked to compare multiple ML models, perform cross-validation, evaluate metrics, or select the best model for a classification/regression task.
Model family compatibility matrix — loaders, resolutions, samplers, CFG, VAE, ControlNet, and LoRA compatibility for SD 1.5, SDXL, Flux, SD3, and video models
Checks whether each data model and field from design docs, schema, DTOs, and APIs has corresponding frontend pages, components, interaction states, validation coverage, and…
AI model configuration patterns for this project. Single source of truth for model definitions. Triggers on "MODEL_CONFIG", "model", "pricing", "capabilities", "auto router",…
自定义大模型自动接入工程师（宿主无关版）。当用户说"接入自定义大模型""配置自己的大模型""把模型接入本Agent""接入工程师""接一个自己的模型"或希望把第三方/自建 OpenAI 兼容模型（DeepSeek、智谱 GLM、Kimi、硅基流动等）自动接入当前 Agent 时使用。触发后 AI…
Universal model detection and capability assessment for optimal cross-model compatibility
The general model-building method for ANY market — framing, sample-size discipline, feature selection (user priors + data-driven), baselines, leakage-safe validation, calibration,…
MCMC diagnostics for Bayesian models including convergence assessment, effective sample size, divergences, and posterior predictive checks.
Generate an ER diagram of Eloquent models in Mermaid format.
Fetch current model names from AI providers (Anthropic, OpenAI, Gemini, Ollama), classify them into tiers (fast/default/heavy), and detect new models.
Use when modeling one Rust domain concept with newtypes, enums, or typestate to prevent illegal states.
Генерация доменных моделей в DSL-формате. Использовать когда пользователь просит создать модель, описать сущность, добавить поля к модели.
Recommend the cost-optimal Claude model (Fable/Opus/Sonnet/Haiku) and effort level for a given task, and delegate to cheap subagents (sonnet-implementer / haiku-scout) when…
Combine diverse models through averaging, blending, and stacking to beat any single model. Use when squeezing maximum accuracy from a competition or high-stakes prediction, after…
Automatic model switching based on task complexity for OpenClaw agents. Dynamically escalates between Haiku (simple), Sonnet (medium), and Opus (complex) models.
Evaluate trained model performance through benchmarking, detect regressions against baseline metrics, and validate checkpoint quality.
Model Evaluation Metrics - Auto-activating skill for ML Training. Triggers on: model evaluation metrics, model evaluation metrics Part of the ML Training skill category.
Model Explainability Tool - Auto-activating skill for ML Training. Triggers on: model explainability tool, model explainability tool Part of the ML Training skill category.
Use when exploring Honeydew semantic layer, discovering entities/fields, setting up workspace and branch context, or running simple structured queries to inspect data.
Use to export trained models — torch.export, ONNX, Core ML, ExecuTorch, TensorRT, parity.
Guide pour le fine-tuning de modèles ML/LLM (LoRA, QLoRA, PEFT, datasets, hyperparamètres) — from general/general-misc
Enforces all 7 Laws as a standing stance — go with Claude Code and the model, not against it. Skills are scaffolding that merges into the model over time; the durable core is…
Model a capacity-tied hiring plan — back-solve the funnel pipeline for target hires, flag the leaking stage, size recruiter capacity, and hand off the comp envelope.
Use when you have MS/MS spectra in .msp format and need to retrieve similar compounds or compute spectral similarities for compound identification.
Use when adding a new model or pipeline to diffusers, setting up file structure for a new model, converting a pipeline to modular format, or converting weights for a new version…
CLI commands to add, update, list, and deprecate AI models in the database. Triggers on "add model", "new model", "update model", "deprecate model", "model pricing", "model cli".
Test, validate, and add new AI models to the eval suite. Use when user asks to add new models, test model access, check pricing, or update models.yml. — from sunholo-data/ailang
Test, validate, and add new AI models to the eval suite. Use when user asks to add new models, test model access, check pricing, or update models.yml.
Diskrete oder kontinuierliche Markov-Ketten erstellen und analysieren, einschliesslich Uebergangsmatrix-Konstruktion, Zustandsklassifikation, Berechnung stationaerer Verteilungen…
Use when when you need to programmatically interface with a TensorFlow Serving model instance and must discover or validate the expected input names (e.
Use when when preparing to send peak data (1H and 13C NMR measurements) to a machine learning classification endpoint and you need to verify the current model's input/output names…
Use when before submitting peak data or other inputs to a machine learning classification API for the first time, after a model update, or if you encounter unexpected prediction…
Plan the migration of an LLM feature from one model to another without breaking production. Use when a model is being deprecated, a newer model looks better or cheaper, or when…
Break down Claude Code usage by model family (Opus / Sonnet / Haiku) from the Agent Monitor dashboard — each family's share of tokens, share of cost, and the spots where an…
Keep production models honest: monitor input/data drift, prediction/concept drift, and performance decay (when labels arrive); define the retraining trigger up front…
Monitor model performance, detect data drift, concept drift, and anomalies in production using Prometheus, Grafana, and MLflow
Use when reducing model size, improving inference speed, or deploying to edge devices - covers quantization, pruning, knowledge distillation, ONNX export, and TensorRT…
Optimisation de modèles ML pour l'inférence (quantization, pruning, distillation, ONNX) — from general/general-misc
Split a model that will not fit on one GPU across devices with tensor, pipeline, and expert parallelism, choosing each split by its communication cost.
Configure or inspect the in-context model-preference layer — model groups (incl. local/BYO models), per-session slots, and task-class routing — by editing the gitignored extend…