zg (zvec-grep) is a newly open-sourced local search tool that aims to bridge the gap between traditional keyword-based search and semantic understanding. Unlike cloud-based search solutions, zg performs all indexing and retrieval on the local device, ensuring that code and documents never leave the machine. This local-first approach is particularly appealing for developers and AI agents that handle sensitive data. The tool combines the speed and precision of ripgrep with the ability to understand fuzzy or ambiguous queries, making it suitable for a wide range of use cases, from code navigation to document retrieval. As AI agents become more prevalent in development workflows, tools like zg offer a privacy-preserving alternative to sending data to external services. The open-source nature of zg also allows the community to inspect and contribute to its development, fostering trust and innovation.
zg (zvec-grep) is a newly open-sourced local search tool designed for both humans and AI agents. It performs indexing and retrieval entirely on-device, protecting privacy while understanding fuzzy intent and maintaining the precision of ripgrep. This addresses growing concerns about data privacy in AI-assisted workflows.