What is nhentai-doujin-shelf?

This repository describes a Python command line tool called Hentaihunter, built to search and index adult manga content, known as doujinshi, across several existing websites at once instead of browsing each site separately. It presents itself as a metadata and cataloging tool rather than a hosting service, saying it pulls public listing information such as titles, tags, and descriptions from these sites without storing or redistributing the underlying images itself. According to the README, a single command can query multiple source sites together, apply filters like tag combinations, date ranges, and rating thresholds, and export the resulting metadata as JSON, CSV, or a SQLite database for personal archiving. It documents a YAML configuration file for setting per source options such as rate limits, proxy addresses, excluded tags, and preferred language, along with command line examples for searching by tag, searching by artist, downloading a specific item by its site ID, and running scheduled batch jobs. The tool also describes an optional integration with OpenAI and Claude APIs, used to automatically generate tags or short descriptions from cover images when that metadata is missing from a source site, controlled by the user's own API keys and off by default. It lists support for Linux, macOS, and Windows, with experimental support through Termux on Android and no iOS support. The README states this is meant for personal use and research and puts the burden on the user to comply with local laws on adult content, respect each source site's terms of service, and confirm they have the right to access and archive anything they collect. Given the explicit nature of the content involved and that scraping third party sites can violate those sites' terms of service, anyone considering this tool should review those legal and platform rules for their own situation before using it.