semantic-search

Search codebases by natural language and return ranked results with metadata.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/benreceveur/claude-workflow-engine --skill semantic-search-benreceveur
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: semantic-search
Source: https://github.com/benreceveur/claude-workflow-engine/tree/main/skills/semantic-search
Command: npx skills add https://github.com/benreceveur/claude-workflow-engine --skill semantic-search-benreceveur

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables natural language searching of codebases, detecting patterns, finding similar code, and understanding code semantics.

Core Features & Use Cases

  • Natural language search: Find code by description.
  • Semantic similarity: Find similar implementations.
  • Pattern detection & cross-reference: Recognize code patterns and API usage.

Quick Start

Example command to search: "search for database connection functions" in path ./src.

Frequently Asked Questions about semantic-search

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

FAQPage Schema
How do I search code by natural language description instead of syntax?

Natural language code search lets you find functions and patterns by describing what you need rather than writing exact code. This Skill interprets your description semantically and returns ranked results matching the intent across your codebase.

Can I use semantic search to find duplicate or similar code implementations?

Yes. Semantic search detects similar implementations and code duplication by analyzing code relationships and patterns, not just string matching. It identifies functionally equivalent code across your codebase regardless of syntax differences.

What's the best way to discover API usage patterns in a large codebase?

Semantic search recognizes API usage patterns and cross-references by computing code semantics. Describe the API or pattern you're investigating, and the Skill returns structured, ranked results showing where and how it's used.

Do I need to preprocess my Python code before running semantic search?

The Skill parses Python AST internally to understand code structure and semantics. Point it at your source directory, and it handles parsing and analysis—no manual preprocessing required.

How does semantic search differ from basic text search across code?

Text search matches keywords; semantic search understands code meaning and relationships. It returns results ranked by relevance to your intent, capturing similar logic and patterns that text search would miss.

Can semantic search work across multiple files and modules at once?

Yes. Provide your codebase path and the Skill searches across all files, detecting patterns, API usage, and similar implementations while returning structured results with metadata and rankings.