Cove also provides some Signals that aren't offered by any of our Integrations. We found these additional Signals useful at our previous companies, so we decided to build them ourselves!

Custom AI Models

You can build custom AI models with Cove to help you enforce any policy. All we require from you is a policy definition, and we walk you through a 10 minute workflow to produce an AI model specifically designed for that policy.

Once you create a custom AI model, you can build Rules that leverage your new model to make automated enforcement decisions.

Text Signals

We offer a few text-based Signals that give you the flexibility to detect harmful text:

• Exact Matching: Look for exact words or phrases in your content.
• Regular Expression (Regex) Matching: Look for text patterns in your Items using regular expressions.
• Text Variant: We've built an algorithm to detect common variants of strings of text. This is particularly useful to catch bad actors trying to evade your enforcement by using leetspeak, replacing characters, adding punctuation in the middle of words, or other forms of evasion. For example, if you're looking for the word "Hello", we'll detect "h3||0" as a match.
• Language Detection: Given some text, we can detect what language it contains. Often the exact language of the text is difficult to discern - particularly when a piece of text contains more than one language. In these cases, we choose the highest probability language.

Location Signals

If you want to set up Rules that target specific locations, you can do so with our Location Matching Signal. With every Item you send us, you'll need to include a geohash representing the latitude-longitude location of the user who created it. Then you can create Rules that only action on Items created in or around particular locations.