I open-sourced a structured JSON dataset for cataloging ecchi and fanservice anime, especially mainstream non-hentai titles with visible nipple scenes.
What is this?#
This is a dataset for tracking ecchi and fanservice scenes in anime, with a particular focus on whether a title contains visible nipple exposure.
By “ecchi” or “fanservice,” I mean mainstream anime that includes sexually suggestive scenes: bath scenes, changing scenes, swimsuit episodes, exposed breasts, butt shots, and similar material made for the viewer’s gaze. Some titles are censored during TV broadcast with steam, light beams, mysterious shadows, or other holy censorship techniques, and later get less censored versions on Blu-ray, DVD, or certain streaming releases.
That said, this project is not a hentai or porn database. The focus is on commercial Japanese anime that still belongs to the mainstream / non-hentai side of anime culture. In other words: this is the cursed middle ground where TV anime, BD upgrades, and horny metadata meet.
This page explains the project, the inclusion criteria, the data format, the source code, and possible future plans.
Scope#
This dataset mainly covers titles such as:
- mainstream Japanese anime
- non-hentai / non-porn anime
- ecchi anime and fanservice-heavy shows
- anime with visible nipple scenes
- bath scenes, changing scenes, swimsuit episodes, and other suggestive scenes
- TV, Blu-ray, OVA, and streaming versions when the differences are relevant
- titles where episode numbers, character names, and scene categories can be organized as metadata
This list began as a private project on my personal website around November 2021. I slowly collected information from multiple sources and organized it into tables, including anime titles, episode numbers, character names, and scene notes.
I have checked more than 200 anime titles while working on this project, but this kind of information is scattered across old blogs, forums, wikis, imageboards, personal notes, and random comments from people who somehow remember the exact episode where the censorship disappeared. There are still missing entries and errors.
Instead of keeping the whole thing as a private cursed spreadsheet, I decided to publish the data in an open format so other people can inspect, reuse, correct, and build upon it.
What is included?#
The dataset may include information such as:
- anime title
- release year or broadcast year
- relevant episode number
- character name
- scene category
- notes for reference
- image and table data used by the Hugo website
The project includes both the JSON data and a sample Hugo website that renders the data into readable tables. This means you can use the same data to build your own instance or frontend.
This specification describes the ideal direction of the project. The current dataset still contains incomplete entries, inconsistent formatting, and items that need further cleanup.
What is not included?#
This project does not include:
- illegal streaming links
- video files
- torrent links
- download links
- hentai / eroge anime / adult-only anime
- doujin anime
- live-action content
- instructions on where to watch or download the anime
This is a metadata project. It is meant to document titles, episodes, characters, and scene categories. It is not a piracy index, not a streaming guide, and not a file-sharing project.
Online demo#
You can view the current public examples here:
- https://htxacg.cc/posts/uncensored-ecchi-anime-list
- https://htxacg.cc/posts/ecchi-fanservice-anime-list
Source code#
The data and website source code are available on GitHub:
- https://github.com/ivon852/ecchi-fanservice-anime-list-data
- https://github.com/ivon852/ecchi-fanservice-anime-list-website
Due to platform limitations, text data and images are stored separately.
The goal is to make the list easier to maintain collaboratively, while allowing other people to freely reuse the raw data and publish their own versions. For example, the Hugo template can be used to generate a browsable instance from the same JSON files.
User experience#
The current website is a static site built with Hugo. JSON data is rendered into HTML tables, with each decade separated into its own table.
I used to write HTML tables by hand. That was a terrible idea. It worked for a while, then the list grew, and the spreadsheet goblin inside me finally admitted defeat.
The site uses lazy loading for images and collapses older decade tables with the details tag to reduce initial page load. Images are stored in S3-compatible storage with CDN acceleration.
CSS and JavaScript are used to adjust the table layout across different screen sizes. On desktop, images are displayed vertically at full width. On mobile, images can be viewed with horizontal scrolling.
For image format, I currently use WEBP instead of AVIF. AVIF is efficient, but WEBP is still easier to handle across more environments and workflows.
Image notes are stored in the title attribute. On desktop, they can be viewed by hovering with the mouse. On mobile, they can usually be viewed by long-pressing the image.
The current Google Lighthouse scores are:
- Performance: 93
- Accessibility: 94
- Best Practices: 100
- SEO: 92
Inclusion criteria#
For detailed inclusion criteria, see the specification in the data repository:
The short version is: the project focuses on mainstream Japanese anime with ecchi / fanservice elements, especially works where visible nipple scenes can be documented with episode and character metadata.
Issues and pull requests are welcome#
This niche is surprisingly difficult to document properly. Information is fragmented, old, inconsistent, and often hidden in forgotten forum posts or ancient personal blogs.
Corrections and contributions are welcome through GitHub Issues or Pull Requests, especially for:
- wrong anime titles
- wrong episode numbers
- missing character names
- TV / Blu-ray / streaming version differences
- titles that fit the inclusion criteria
- inconsistent romanization or naming
- JSON structure improvements
- better metadata fields
If you know the exact episode where the censorship lifts, congratulations: you are exactly the kind of dangerous scholar this project needs.
English / Japanese / Chinese versions#
This project originally started from a Taiwanese Mandarin personal website. Since anime fandom is scattered across languages, the project is being prepared for Chinese, Japanese, and English readers.
The goal is to let anime fans, researchers, wiki editors, data hoarders, and terminally online weebs refer to the same structured data instead of rebuilding the same list from scratch over and over again.
Future plans#
The current site is still a Hugo-based static website. It works, but a single huge page full of tables and images will eventually hit limits in editing, loading speed, and maintainability.
Possible future improvements include:
- moving the website to a more robust server environment
- replacing the purely static architecture with a frontend / backend setup
- building a React frontend
- using Node.js and PostgreSQL to manage the data
- storing images in S3-compatible storage
- optimizing images through an image processing pipeline
- creating one page per anime title
- building a browser-based editing interface
- integrating external anime database APIs
- adding a random gallery page
- turning the project into a wiki-like collaborative database
In the long run, I want this to grow from a personal list into reusable open data for documenting ecchi and fanservice scenes in mainstream anime. It is niche. It is weird. It is probably useful to five people and one cursed spreadsheet enjoyer. But that is more than enough reason to make it public.
