SuperAPI AI Directory

embark-search

Coding & development

View repository

Updated 10-15 23:30

About this skill

Semantic Code Search

Use embark search to find code snippets by meaning, not just keywords. If it's not installed use embark-install skill.

Use it as a single semantic bootstrap when the relevant file or subsystem is unknown. Do one broad search, open and inspect at least one returned file locally, and inspect nearby code in that same directory or subsystem before any retry. If that still does not identify the needed adjacent area, do a narrowed retry with embark search -p <path> ... using the directory of the best first hit.

Usage

embark search "<detailed and descriptive query>"
embark search "<code snippet>" # find similar snippets
embark search -p <path> "<query>"  # <path> must be relative to the project root

Query Tips

  • Be descriptive: "function that validates user email addresses" > "email"
  • Include context: "error handling middleware for HTTP requests with logging"
  • Specify what you're looking for: "React component that renders a modal dialog"
  • Make the first query specific to the issue's named feature, class, method, or behavior when available

Examples

# Find authentication-related code
embark search "user authentication login flow"

# Narrow to specific directory
embark search -p src/auth "JWT token validation"