What problem does it solve? AI coding agents often waste context window capacity by loading entire files when only a few lines are relevant. This Skill enforces a search-first discipline so agents locate code with minimal file reads. ## Core Features & Use Cases - Search-First Workflow: Defines a five-step process from question definition through keyword extraction, codebase search, result evaluation, and targeted reading. - Cross-Platform Search Commands: Provides ready-to-use ripgrep, grep, and PowerShell Select-String patterns for locating identifiers. - Efficiency Metrics & Anti-Patterns: Includes measurable criteria (files searched vs. fully read, lines read) and concrete good/bad examples to keep context lean. - Use Case: Before fixing a bug in an unfamiliar repository, search for the failing function name, read only the matching 20-line section, and report candidate files instead of loading entire modules. ## Quick Start Ask the agent to find where a specific function or endpoint is defined using search-first context fetching before reading any files.
