Show HN: Pviz-parser – codebase parsing package for Python and TS/JS codebases (github.com)

A new tool called pviz-parser has been released, designed to generate LLM-optimized dependency analysis bundles for Python and TypeScript/JavaScript projects. This package analyzes a project's dependency graph and creates a structured JSON bundle that captures comprehensive details about the codebase, including nodes, edges, import relationships, cycle detection, and metrics per file. Developers can easily integrate this tool into their workflow using a simple command-line interface, enabling them to present their entire codebase's structure to language models like ChatGPT or Claude in a single, compact artifact. The significance of pviz-parser lies in its ability to facilitate more effective interaction with large language models (LLMs) by providing them with a complete contextual overview of coding projects. Instead of engaging with files one by one, users can upload a single bundle that represents their entire codebase, enhancing the potential for insightful analysis and troubleshooting. Supported by Python, TypeScript, and JavaScript, with partial support for Java, and the capability to handle polyglot repositories, pviz-parser is positioned as a valuable asset for developers looking to leverage AI in their coding processes. The tool is open source and runs locally, providing complete ownership of generated bundles, while a hosted version offers broader language support and additional features.