{"slug":"datarobot-feature-engineering","title":"datarobot-feature-engineering","summary":"Guidance for feature engineering, feature discovery, feature importance analysis, and understanding DataRobot's automated feature engineering capabilities. Use when working with feature engineering, feature discovery, or analyzing feature importance in DataRobot.","platform":"Claude","tags":[],"authorName":"LLM Mart","authorSlug":"llm-mart","score":0,"source":"github","price":null,"verified":false,"createdAt":"2026-09-11T17:36:44.72129Z","repo":{"url":"https://github.com/datarobot-oss/datarobot-agent-skills","stars":27,"forks":23,"license":"Apache-2.0","updatedAt":"2026-09-24T02:43:43Z"},"bodyHtml":"<hr>\n<h2>name: datarobot-feature-engineering\ndescription: Guidance for feature engineering, feature discovery, feature importance analysis, and understanding DataRobot's automated feature engineering capabilities. Use when working with feature engineering, feature discovery, or analyzing feature importance in DataRobot.</h2>\n<h1>DataRobot Feature Engineering Skill</h1>\n<p>This skill provides guidance for working with features in DataRobot, including understanding automated feature engineering, analyzing feature importance, and optimizing feature sets.</p>\n<h2>Quick Start</h2>\n<p><strong>Most common use case</strong>: Analyze feature importance for a model</p>\n<ol>\n<li><strong>Get feature importance</strong>: <code>get_feature_importance(model_id)</code> to get importance scores</li>\n<li><strong>Analyze top features</strong>: Sort by importance and identify key drivers</li>\n<li><strong>Export feature list</strong>: <code>export_feature_list(project_id)</code> to document features</li>\n</ol>\n<p><strong>Example</strong>: \"Show me the top 10 most important features for model xyz123\"</p>\n<h2>When to use this skill</h2>\n<p>Use this skill when you need to:</p>\n<ul>\n<li>Understand what features DataRobot creates automatically</li>\n<li>Analyze feature importance for models</li>\n<li>Discover which features drive predictions</li>\n<li>Optimize feature sets for better performance</li>\n<li>Understand feature types and transformations</li>\n<li>Export feature lists and definitions</li>\n</ul>\n<h2>Key capabilities</h2>\n<h3>1. Feature Discovery</h3>\n<ul>\n<li>Understand automated feature engineering in DataRobot</li>\n<li>Review derived features and transformations</li>\n<li>Identify feature types (numeric, categorical, text, date)</li>\n<li>Explore feature relationships and interactions</li>\n</ul>\n<h3>2. Feature Importance Analysis</h3>\n<ul>\n<li>Get feature importance scores for models</li>\n<li>Understand which features drive predictions</li>\n<li>Compare feature importance across models</li>\n<li>Identify redundant or low-value features</li>\n</ul>\n<h3>3. Feature Optimization</h3>\n<ul>\n<li>Select important features for model performance</li>\n<li>Remove low-importance features to reduce complexity</li>\n<li>Understand feature impact on predictions</li>\n<li>Optimize feature sets for deployment</li>\n</ul>\n<h3>4. Feature Documentation</h3>\n<ul>\n<li>Export feature lists and definitions</li>\n<li>Document feature transformations</li>\n<li>Understand feature derivation logic</li>\n<li>Share feature information with stakeholders</li>\n</ul>\n<h2>Workflow examples</h2>\n<h3>Example 1: Analyze feature importance</h3>\n<p><strong>User request</strong>: \"Show me the top 10 most important features for model xyz123 and explain what they mean.\"</p>\n<p><strong>Agent workflow</strong>:</p>\n<ol>\n<li>Get feature importance scores for the model</li>\n<li>Sort features by importance (descending)</li>\n<li>Get top 10 features with their scores</li>\n<li>Retrieve feature metadata and descriptions</li>\n<li>Explain what each feature represents and why it's important</li>\n<li>Provide insights on feature relationships</li>\n</ol>\n<h3>Example 2: Optimize feature set for deployment</h3>\n<p><strong>User request</strong>: \"Create a simplified feature set for deployment abc123, keeping only features with importance &gt; 0.1.\"</p>\n<p><strong>Agent workflow</strong>:</p>\n<ol>\n<li>Get feature importance for the deployed model</li>\n<li>Filter features by importance threshold (&gt; 0.1)</li>\n<li>Verify filtered features are sufficient for predictions</li>\n<li>Document the optimized feature set</li>\n<li>Update deployment configuration if needed</li>\n</ol>\n<h2>Using DataRobot SDK</h2>\n<p>This skill guides you to use the DataRobot Python SDK directly. Install the SDK if needed:</p>\n<pre><code>pip install datarobot\n</code></pre>\n<h3>Key SDK Operations</h3>\n<p>Use these DataRobot SDK methods for feature analysis:</p>\n<p><strong>Feature Information</strong>:</p>\n<ul>\n<li><code>model.get_features()</code> - List all features in a model</li>\n<li><code>model.get_feature_impact()</code> - Get feature importance scores</li>\n<li><code>project.get_features()</code> - List features in a project</li>\n</ul>\n<p><strong>Feature Analysis</strong>:</p>\n<ul>\n<li><code>feature.name</code> - Feature name</li>\n<li><code>feature.feature_type</code> - Feature type (Numeric, Categorical, etc.)</li>\n<li><code>feature.importance</code> - Feature importance score</li>\n</ul>\n<p>See the <a href=\"#common-patterns\">Common Patterns</a> section below for complete examples.</p>\n<h2>Best practices</h2>\n<ol>\n<li><strong>Review automated features</strong>: DataRobot creates many derived features automatically - review them</li>\n<li><strong>Focus on important features</strong>: Pay attention to high-importance features for insights</li>\n<li><strong>Understand feature types</strong>: Different feature types require different handling</li>\n<li><strong>Feature documentation</strong>: Document important features for stakeholders</li>\n<li><strong>Feature selection</strong>: Consider removing very low-importance features for simplicity</li>\n<li><strong>Feature stability</strong>: Consider feature stability over time, not just importance</li>\n</ol>\n<h2>Common patterns</h2>\n<h3>Pattern 1: Feature importance analysis</h3>\n<pre><code>import datarobot as dr\n\n# Initialize client\ndr.Client()\n\n# Get model and feature importance\nmodel = dr.Model.get(\"xyz123\")\nfeature_impact = model.get_feature_impact()\n\n# Sort by importance\nsorted_features = sorted(\n    feature_impact, key=lambda x: x.get(\"impactNormalized\", 0), reverse=True\n)\n\n# Get top 10 features\ntop_features = sorted_features[:10]\nfor feature in top_features:\n    print(f\"{feature['featureName']}: {feature.get('impactNormalized', 0):.3f}\")\n</code></pre>\n<h3>Pattern 2: Feature filtering</h3>\n<pre><code>import datarobot as dr\n\n# Get model and feature importance\nmodel = dr.Model.get(\"xyz123\")\nfeature_impact = model.get_feature_impact()\n\n# Filter by importance threshold (&gt; 0.1)\nimportant_features = [f for f in feature_impact if f.get(\"impactNormalized\", 0) &gt; 0.1]\n\nprint(f\"Found {len(important_features)} features with importance &gt; 0.1\")\n</code></pre>\n<h2>Feature types in DataRobot</h2>\n<h3>Numeric Features</h3>\n<ul>\n<li>Continuous numeric values</li>\n<li>Automatically scaled and normalized</li>\n<li>Can be used in mathematical operations</li>\n</ul>\n<h3>Categorical Features</h3>\n<ul>\n<li>Discrete categories or labels</li>\n<li>Automatically encoded (one-hot, target encoding)</li>\n<li>Important for many model types</li>\n</ul>\n<h3>Text Features</h3>\n<ul>\n<li>Text data (descriptions, comments)</li>\n<li>Automatically processed with NLP techniques</li>\n<li>Creates multiple text-derived features</li>\n</ul>\n<h3>Date/Time Features</h3>\n<ul>\n<li>Temporal data</li>\n<li>Automatically creates time-based features</li>\n<li>Important for time series models</li>\n</ul>\n<h2>Understanding feature importance</h2>\n<p>Feature importance scores indicate:</p>\n<ul>\n<li><strong>High importance (&gt; 0.1)</strong>: Feature significantly impacts predictions</li>\n<li><strong>Medium importance (0.05-0.1)</strong>: Feature contributes to predictions</li>\n<li><strong>Low importance (&lt; 0.05)</strong>: Feature has minimal impact</li>\n</ul>\n<p>Note: Importance thresholds vary by model type and problem domain.</p>\n<h2>Error handling</h2>\n<p>Common errors and solutions:</p>\n<ul>\n<li><strong>Feature not found</strong>: Verify feature name and model compatibility</li>\n<li><strong>Importance unavailable</strong>: Some model types don't provide importance scores</li>\n<li><strong>Feature access errors</strong>: Check project and model permissions</li>\n</ul>\n<h2>SDK Setup</h2>\n<h3>Install DataRobot SDK</h3>\n<pre><code>pip install datarobot\n</code></pre>\n<h3>Initialize Client</h3>\n<pre><code>import datarobot as dr\n\ndr.Client()\n</code></pre>\n<h2>Resources</h2>\n<ul>\n<li><a href=\"https://datarobot-public-api-client.readthedocs-hosted.com/\">DataRobot Python SDK Documentation</a></li>\n<li><a href=\"https://docs.datarobot.com/en/docs/modeling/index.html\">DataRobot Feature Engineering Documentation</a></li>\n<li><a href=\"https://docs.datarobot.com/en/docs/modeling/analyze-models/index.html\">Feature Importance Guide</a></li>\n<li><a href=\"https://docs.datarobot.com/en/docs/data/transform-data/feature-discovery/index.html\">Feature Discovery Documentation</a></li>\n</ul>\n","files":[{"path":"SKILL.md","sizeBytes":6999,"isText":true}],"reviewScore":null,"reviewSummary":null,"trust":{"provenance":"trusted-source-unreviewed","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow.","bodySource":null},"bodyLocked":false,"purchaseUrl":null,"sourceUrl":null,"report":{"provenance":"trusted-source-unreviewed","screen":{"ran":true,"outcome":"clean","suspicious":0,"notes":0,"hiddenCharacters":false},"virusScan":{"engine":"clamav","status":"clean","scannedAt":"2026-09-11T17:37:01.265982Z","sha256":"3EE40EF7470D88B03BB46C5AD5F9E70D9ABAF304632CF2C00D213759559A327B","sizeBytes":2485},"review":null,"source":{"repositoryUrl":"https://github.com/datarobot-oss/datarobot-agent-skills","path":"skills/datarobot-feature-engineering","license":"Apache-2.0","commit":"023e5b77fb4c3f651f52d9afc943d87061e22769","subtreeSha":"4F809607FB46FADCE12788AA6A9A1B8306F22F541DDEAF70B367CF96692B81F1","lastSyncedAt":"2026-09-24T06:49:20.424986Z"},"reviewedAt":"2026-09-11T17:40:12.593226Z","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow."},"install":[{"target":"skills-cli","command":"npx skills add https://github.com/datarobot-oss/datarobot-agent-skills/tree/main/skills/datarobot-feature-engineering"},{"target":"claude-code","command":"claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install datarobot-oss-datarobot-agent-skills@llmmart"},{"target":"git","command":"git clone https://github.com/datarobot-oss/datarobot-agent-skills.git"}]}