search-code

Search codebases using tiered semantic and literal search strategies.

8|3|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/lbruton/StakTrakr --skill search-code-lbruton
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search-code
Source: https://github.com/lbruton/StakTrakr/tree/main/.claude/skills/search-code
Command: npx skills add https://github.com/lbruton/StakTrakr --skill search-code-lbruton

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured, tiered approach to searching codebases, minimizing token costs and latency while maximizing accuracy for finding functions, patterns, and understanding code architecture.

Core Features & Use Cases

  • Tiered Search Strategy: Prioritizes semantic search (Claude-Context) before falling back to literal search (Grep/Glob) and then comprehensive investigation (Explore agent).
  • Cost Optimization: Designed to reduce token usage and improve response times.
  • Use Case: When trying to understand how a specific feature like user authentication is implemented across a large project, this Skill guides you through the most efficient search methods.

Quick Start

Use the search-code skill to find all JavaScript functions related to fetching and caching spot prices.

Frequently Asked Questions about search-code

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What is the best way to search a large codebase for specific feature implementations?

A tiered code search strategy minimizes token costs and latency by starting with semantic search via Claude-Context, then falling back to literal string matching with Grep or Glob before triggering a comprehensive Explore agent investigation.

How do I find JavaScript functions related to fetching and caching data across a project?

To find JavaScript functions for fetching and caching data, apply a tiered search approach that begins with semantic search to understand the architecture, then uses literal Grep matching to locate the exact function definitions and patterns.

How does semantic code search compare to using standard grep for codebase navigation?

Semantic code search via Claude-Context understands architectural context and feature relationships, whereas standard Grep only provides literal string matching, making a tiered approach more efficient for balancing accuracy, speed, and token cost.

Can I use the Explore agent for comprehensive codebase investigation when literal search fails?

Yes, the Explore agent serves as the final tier for comprehensive codebase investigation, triggered when semantic search via Claude-Context and literal string matching with Grep or Glob are insufficient to locate specific code implementations.

When should I use Grep instead of semantic search to find code patterns?

Use Grep for literal string matching instead of semantic search when you know the exact syntax or variable names, optimizing for speed and lower token cost during codebase navigation before escalating to the Explore agent.