Is it compatible with both Linux and macOS?

Yes, the setup skill automatically detects your operating system and installs the appropriate binaries and ONNX Runtime libraries for macOS and Linux.

Codixing is a high-performance code retrieval engine designed specifically for AI agents, featuring AST parsing, hybrid BM25+vector search, and dependency mapping.

Does this skill support vector embeddings?

Yes, you can enable semantic search by passing the --embeddings flag or specifying a supported model like bge-small-en or all-minilm-l6-v2.

How do I verify the installation was successful?

The skill includes a verification step where you can run 'codixing search' to test the index and ensure the engine is returning relevant code results.

Codixing Code Retrieval Setup

Name: Codixing Code Retrieval Setup
Author: ferax564

byferax564

•

开发者工具

Initializes and configures the ultra-fast Codixing search engine to optimize codebase discovery and context retrieval.

Codixing-setup automates the integration of the Codixing retrieval engine into your development workflow, enabling advanced semantic search and codebase analysis for AI agents. It handles the binary installation, project indexing via hybrid BM25 and vector search, and dependency graph generation, providing Claude with a high-performance map of your project structure for significantly more accurate code navigation and contextual awareness.

主要功能

011 GitHub stars

02Hybrid search indexing using BM25 and optional vector embeddings

03Automated installation and version checking of the Codixing engine

04Support for multiple embedding models including bge-small-en and all-minilm

05AST-based parsing for precise symbol and dependency lookups

06Architecture overview generation with codebase mapping

使用场景

01Optimizing AI agent context retrieval for large-scale legacy codebases

02Enhancing project navigation through dependency graph and symbol mapping

03Bootstrapping a new project with advanced AI-ready code search capabilities

主要功能

011 GitHub stars

02Hybrid search indexing using BM25 and optional vector embeddings

03Automated installation and version checking of the Codixing engine

04Support for multiple embedding models including bge-small-en and all-minilm

05AST-based parsing for precise symbol and dependency lookups

06Architecture overview generation with codebase mapping

使用场景

01Optimizing AI agent context retrieval for large-scale legacy codebases

02Enhancing project navigation through dependency graph and symbol mapping

03Bootstrapping a new project with advanced AI-ready code search capabilities