productivity▌

Credit risk data cleaning and variable screening pipeline for pre-loan modeling. \n \n Executes 11 independent steps covering data loading, abnormal period filtering, missing rate analysis, low-IV and high-PSI variable removal, null importance denoising, and correlation-based feature elimination \n Supports organization-level analysis with separate modeling and out-of-sample (OOS) sample handling, plus multi-process acceleration for IV and PSI calculations \n Generates comprehensive Excel report

Review Google Meet attendance records and participant session durations. \n \n Requires the gws-meet skill to access Google Meet conference data \n Three-step workflow: list recent conferences, retrieve participant details, and view individual session timestamps and durations \n Returns tabular output for easy review of who attended and how long they stayed \n

Parse, filter, deduplicate, and aggregate SARIF files from static analysis tools. \n \n Reads and processes SARIF 2.1.0 output from CodeQL, Semgrep, and other scanners; does not run scans itself \n Supports filtering by severity, extracting findings by file or rule, and deduplicating alerts using fingerprints and partial fingerprints \n Provides three strategies: jq for quick CLI queries, pysarif for programmatic object access, and sarif-tools for aggregation and format conversion \n Handles pat

Systematic approach to measuring, analyzing, and improving website loading speed and Core Web Vitals. \n \n Covers five-step optimization workflow: establish baseline metrics with Lighthouse, identify bottlenecks (bundle size, images, render-blocking resources), prioritize high-impact fixes, implement improvements, and verify results \n Addresses Core Web Vitals (LCP, FID, CLS) with specific code examples for image optimization, code splitting, lazy loading, and layout shift prevention \n Includ

Agents on Eve Horizon have no built-in "memory" primitive, but the platform provides storage systems at every timescale. This skill teaches how to compose them into coherent memory for agents that learn, remember, and share.

ln-652-transaction-correctness-auditor

Localize VS Code extensions across configurations, walkthrough content, and source code messages. \n \n Three localization approaches: package.nls.LANGID.json for settings, commands, menus, and views; markdown files for walkthrough content; bundle.l10n.LANGID.json for source code strings \n Covers all user-facing resources including contributed configurations, commands, menus, views, ViewsWelcome, and walkthrough titles and descriptions \n Requires creating or updating localization files for all

READMEs answer questions your audience will have. Different audiences need different information - a contributor to an OSS project needs different context than future-you opening a config folder.

Turn a user request into a single, actionable plan with atomic steps.

Create GitHub issues from specification files using the feature_request.yml template. \n \n Analyzes specification files to extract requirements and automatically generates issue titles and descriptions \n Searches existing issues to prevent duplicates before creating new ones \n Supports both creating new issues and updating existing ones via the GitHub API \n Applies feature and enhancement labels automatically based on specification content \n