vectorize

Manage codebase and database vectorization for semantic search.

Updated Mar 11, 2026
One-click install
npx skills add https://github.com/mdmagnuson-creator/helm-ade-toolkit --skill vectorize
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: vectorize
Source: https://github.com/mdmagnuson-creator/helm-ade-toolkit/tree/main/skills/vectorize
Command: npx skills add https://github.com/mdmagnuson-creator/helm-ade-toolkit --skill vectorize

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @anthropic-ai/sdk, @lancedb/lancedb, apache-arrow, commander, glob, mysql2, openai, ora, pg, tree-sitter, tree-sitter-go, tree-sitter-java, tree-sitter-javascript, tree-sitter-python, tree-sitter-rust, tree-sitter-typescript, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables semantic search across your entire codebase and database, allowing you to find information using natural language queries instead of keyword matching.

Core Features & Use Cases

  • Semantic Code Search: Ask questions like "How does user authentication work?" and get relevant code snippets.
  • Database Schema Indexing: Understand your database schema and configuration tables semantically.
  • Use Case: When refactoring, ask "Show me all functions related to payment processing" to quickly find relevant code and understand its context.

Quick Start

Use the vectorize skill to initialize vectorization for the current project.

Frequently Asked Questions about vectorize

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

FAQPage Schema
How do I perform semantic search across my codebase using natural language?

You can index your codebase for semantic search by initializing vectorization, which uses AST parsing to chunk code and embedding APIs like OpenAI to generate vectors. This enables natural language queries to find relevant code snippets.

Can I index my database schema for semantic search?

Yes, the Skill supports database schema and config table indexing. It vectorizes your database structure, enabling semantic retrieval to understand table relationships and configurations using natural language questions.

Does semantic search work with JavaScript, Python, Go, Rust, and Java codebases?

Yes, semantic search supports JavaScript, Python, Go, Rust, Java, and TypeScript codebases. It uses tree-sitter parsers for these languages to perform AST parsing, ensuring accurate code chunking before generating embeddings.

Do I need an OpenAI API key to vectorize my code?

You need an embedding API key to vectorize code. The Skill integrates with OpenAI and Voyage AI APIs to generate embeddings from code chunks, which are then stored in a LanceDB vector database for retrieval.

What is the best way to find functions related to a specific feature during refactoring?

The best way to find related functions during refactoring is querying your vectorized codebase index. By asking natural language questions about the feature, you retrieve relevant code snippets and their context without relying on exact keyword matches.