(Pending Approval From Slack)

This is a Slack bot that uses machine learning to infer if anything posted to Slack is NSFW.

There are two core services involved

• Safely Listener, The Real Time Slack events processor that dispatches classification tasks.
• Safely Moderator, The Classifier that receives tasks from the real time listener and makes inferences about content.

If you don't want to bother deploying this and running your own infrastructure, you can join Safely (coming soon), which is a paid content moderation service that provides the same capability in addition to:

• Fine tuning inference parameters.
• Continuously trained classification models.
• Custom reporting and behaviors.
• Additional content-moderation insights.
• Daily/Weekly digests.

If you prefer to run this yourself, follow these initial steps

• Create a Slack App and Bot app in Slack.
• export your Slack token as SAFELY_SLACKBOT_TOKEN=<your token>.
• export your list of Slack admins as SAFELY_ADMINS_LIST=@michael,@admin (separated by a comma).
• Invite your Slack bot to all the channels you would like to monitor.

• Create a new cloud compute instance with at least 4GB RAM and 4 vCPUs.

• Clone this repo.

• Install Docker and Docker Compose.

• Start the entire stack with

    docker-compose up -d
  

To run with just Docker:

Run a Celery Broker (Redis or RabbitMQ)

docker run --rm --name redis -d redis

Start the events listener service

docker run --rm --name listener --link redis:redis --env-file ./.env -d imichael/safely-listener

Start the moderator service

docker run --rm --name moderator --link redis:redis --env-file .env -d imichael/safely-moderator

See docker-compose.yml for supported env vars to store in the .env file.

The k8s.yaml manifest will provision:

• A safely-listener Deployment
• A safely-moderator Deployment
• A safely-redis Deployment
  • An instance of redis used for communication by the various celery components
• A safely-redis Service so that the redis instance can be accessed by hostname
• A safely-config Configmap
• A safely-slackbot-token Secret

Assuming you have the slackbot token already saved in a file, you can deploy with the command below, by specifying the path to your token file.

sed "s/SLACKBOT_TOKEN_PLACEHOLDER/$(cat PATH/TO/FILE/WITH/TOKEN)/" k8s.yaml | kubectl apply -f -

• Safely focuses on basic NSFW content including nudity and profanity. See the Open NSFW project for further information about the scope of the model.

• This tool is imperfect. There will be some false positives & negatives. Please see the license file to learn more about guarantees (there are none).

Special thanks to:

• Yahoo's NSFW Model, which has been converted from Caffe to Tensorflow.

• Profanity Check, which provides a model trained by Sci Kit learn (not a black list)

MIT