Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/Namchee/pleasantcord
🤖 NSFW image moderation bot for Discord
https://github.com/Namchee/pleasantcord
bot discord nsfw-detection
Last synced: 6 days ago
JSON representation
🤖 NSFW image moderation bot for Discord
- Host: GitHub
- URL: https://github.com/Namchee/pleasantcord
- Owner: Namchee
- License: mit
- Archived: true
- Created: 2020-09-02T07:42:14.000Z (about 4 years ago)
- Default Branch: master
- Last Pushed: 2023-10-11T15:08:21.000Z (about 1 year ago)
- Last Synced: 2024-08-01T21:43:50.277Z (3 months ago)
- Topics: bot, discord, nsfw-detection
- Language: TypeScript
- Homepage: https://pleasantcord.namchee.dev
- Size: 18.9 MB
- Stars: 7
- Watchers: 3
- Forks: 2
- Open Issues: 5
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-made-by-indonesian - Pleasantcord - `NSFW image moderation bot for Discord` *by [Cristopher](https://github.com/Namchee)* (P)
- made-in-indonesia - Pleasantcord - `NSFW image moderation bot for Discord` *by [Cristopher](https://github.com/Namchee)* (P)
README
# Pleasantcord
> This bot has been deprecated. Do not try to visit the dashboard or invite the bot!
Pleasantcord is a simple NSFW image auto moderation bot 🤖 for Discord powered by [TensorFlow JS](https://www.npmjs.com/package/@tensorflow/tfjs-node). On the inside, it uses a pretrained model provided by [nsfw.js](https://github.com/infinitered/nsfwjs) that is able to distinguish image category to several categories.
> This is just a demo. Of course Pleasantcord won't remove Cuphead images!
## Installation
Invite this bot to your server by clicking [this link](https://discord.com/api/oauth2/authorize?client_id=750668307555942482&permissions=2147493888&scope=bot%20applications.commands).
The bot itself requires the following permission in your server, make sure that the bot has **all** the permissions below:
- `Send Messages`
- `Manage Messages`
- `Use Slash Commands`Besides the above permissions, make sure that `pleasantcord` is able to view and access your channels.
Failed to comply with those requirements may result `pleasantcord` to throw `Missing permissions` or `Missing Access`.
## Commands
All commands are prefixed with `pc!`
Command | Description
------- | -----------
`status` | Show the bot status (ping, environment, etc).
`config` | View server config
`help` | Show the bot help menu.## FAQ
### What's the default behavior of `pleasantcord`?
By default, `pleasantcord` will classify any image contents as NSFW when it has `Hentai` or `Porn` label with more than 70% accuracy and delete it from any **text channel** that isn't configured to be a NSFW channel.
### Animated `webp` and `png` throw errors!
Discord cannot display them correctly, yet there's no way to distinguish them from normal non-animated one as they share the same header.
### The classification is incorrect!
I'm not the one who created the model, so I can't say anything about it. That being said, image classification is an ML algorithm which could produce false alarms from time to time, so false alarms are quite common occurence.
## Data Privacy
Technically, `pleasantcord` won't store any of your data. However when `pleasantcord` has encountered an error on your server, `pleasantcord` will store the stacktrace on the logging system which *may* include your personal data (which is 100% embeddable content). This is done to ensure easier error debugging and fixing.
When the issue has been detected, acknowledged, and analyzed, your personal data will **immediately** deleted. If you feel that your data is still stored on our system, ~~feel free to join our support server and contact the moderators directly.~~
## Dashboard
You can access the full open-source code of the dashboard [here](https://github.com/Namchee/pleasantcord-dashboard)
## Credits
- [infinitered](https://github.com/infinitered) — Provides an easy-to-use pretrained model for NSFW detection.
- [Namchee](https://github.com/Namchee/pleasantcord) — (me) For working on the project!## License
This project is licensed under the [MIT License](LICENSE)