Loading...
Loading...
Found 1,899 Skills
Convert evaluation traces and production logs into SFT examples and preference pairs. Use when graded traces or failure examples exist and need to become training data, when applying rejection sampling to model outputs, or when building DPO pairs from passing and failing runs.
Generate and critically evaluate grounded improvement ideas for the current project. Use when asking what to improve, requesting idea generation, exploring surprising improvements, or wanting the AI to proactively suggest strong project directions before brainstorming one in depth. Triggers on phrases like 'what should I improve', 'give me ideas', 'ideate on this project', 'surprise me with improvements', 'what would you change', or any request for AI-generated project improvement suggestions rather than refining the user's own idea.
Selects, deploys, and customizes AI models on Amazon SageMaker. Fine-tuning (SFT, DPO, RLVR, RLAIF), model selection, dataset preparation, evaluation, deployment to SageMaker endpoints or Bedrock, and endpoint diagnostics. Covers the full lifecycle from planning through production. Use when fine-tuning models on SageMaker, selecting base models from SageMaker Hub, finding a model to deploy without fine-tuning, transforming datasets for training, checking data readiness, evaluating model quality, deploying to endpoints, setting up IAM roles and S3 buckets for training jobs, or managing a SageMaker Managed MLflow app. Also use to check endpoint health, diagnose failures, debug latency or errors, or view container logs and CloudWatch metrics. Covers Serverless Model Customization, Nova and OSS deployment paths, and PySDK v3 usage. NOT for Ground Truth labeling, Feature Store, or general-purpose AWS infrastructure.
MSBuild property definition patterns: conditional defaults, composition/concatenation, path normalization, trailing-slash handling, TFM detection helpers, and evaluation order. USE FOR: diagnosing and fixing property definition issues and shared-property anti-patterns in .props/.csproj; DefineConstants or NoWarn overwritten instead of appended; unconditional assignments that block project-level overrides; unquoted conditions that fail on empty properties; hardcoded paths that break cross-platform builds; setting overridable defaults; property evaluation order and last-write-wins semantics. DO NOT USE FOR: props vs targets placement (use directory-build-organization), item operations (use item-management), target structure (use target-authoring), general anti-patterns (use msbuild-antipatterns), non-MSBuild build systems.
Design or review load-bearing architecture: authority, ownership, public, persisted, or cross-boundary contracts, dependency boundaries, recovery, architecture guards, migration, rewrite, deprecation, deletion, and drift. Use when work establishes, changes, or evaluates these decisions or controls; skip focused work that only consumes them as supplied inputs and preserves them.
Test for regulatory compliance: GDPR/CMP consent verification, Google Consent Mode v2, Global Privacy Control (GPC), CCPA/US state opt-out, EU AI Act Article 50 transparency, Better Ads Standards, and cookie-inventory auditing. Covers automated consent-flow testing, third-party script blocking before consent, and cookie drift detection. Use when: "GDPR test," "compliance," "CMP test," "cookie consent," "consent mode," "CCPA," "GPC," "AI Act," "Better Ads," "privacy banner." Not for: WCAG/axe-core test authoring — use accessibility-testing. Not for: OWASP/vuln scanning — use security-testing. Not for: evaluating your LLM feature's quality or safety — use ai-system-testing. Related: accessibility-testing, security-testing, ai-system-testing, ci-cd-integration.
Use this skill for any question or action about the user's AI/GenAI applications or agents — their behavior, prompts/responses, quality, hallucinations, guardrails, security, cost/tokens, errors, evaluations/policies, model pricing, or configuration — including comparing or tracking agents over time. It covers both analyzing AI telemetry (GenAI spans) and managing AI Center config via the `cx ai-center` commands.
Train ML models on Databricks. Use for: classification/regression/deep-learning (XGBoost, scikit-learn, LightGBM, PyTorch) with Optuna, @prod/@challenger aliases, batch scoring (spark_udf for plain models, fe.score_batch for feature-store-backed), custom PyFunc, custom ResponsesAgent (LangGraph + UC Function/Vector Search); UC feature tables + FeatureLookup + point-in-time joins + Lakebase online store; declarative Feature Views (create_feature, DeltaTableSource, RollingWindow/SlidingWindow/TumblingWindow, materialize_features, streaming Kafka features). NOT for: endpoint ops (databricks-model-serving), MLflow evaluation (databricks-mlflow-evaluation).
Refactor Scikit-learn and machine learning code to improve maintainability, reproducibility, and adherence to best practices. This skill transforms working ML code into production-ready pipelines that prevent data leakage and ensure reproducible results. It addresses preprocessing outside pipelines, missing random_state parameters, improper cross-validation, and custom transformers not following sklearn API conventions. Implements proper Pipeline and ColumnTransformer patterns, systematic hyperparameter tuning, and appropriate evaluation metrics.
Individual project health assessment skill for evaluating project status across multiple dimensions. Provides early warning of troubled projects.
Run agentlint CLI after code changes to catch patterns for AI evaluation. Activate when finishing code modifications, before committing, or when the developer asks to lint, scan, or review code with agentlint. Covers agentlint check, agentlint list, agentlint review, agentlint init, inline suppression, and output interpretation.
Create professional presentations using the Pyramid Principle methodology. Supports PPTX generation, Marp/Reveal.js Markdown slides, chart creation, speaker notes, and self-evaluation rubrics. Minimal intake form to rapid output workflow.