CPU-friendly local ML

Maskara

Ask AI without exposing people.

A compact PII detection model and SDK for adding privacy middleware to your codebase. It masks sensitive prompt data before cloud AI sees it.

You send Raw prompt inside your app
0/3,000
tiny browser demo
Maskara changes Ready.
0 found
Cloud sees this safe prompt 

          

        


        

        

          

            

              You receive
              Original values restored locally
            

            0 sensitive spans
          

          

            

              App receives this
              restored reply
            

            

          

        

      

    


    

      

        
Model.

        
The public checkpoint lives on Hugging Face. The SDK is built to plug into an app before an LLM call.

      


      

        

          
SDK shape

          
One small wrapper around the prompt path.

          

            
1. ProtectDetect and mask local PII.

            
2. CallSend only safe context.

            
3. RestorePut originals back locally.

          

          
from maskara import Maskara

maskara = Maskara()
safe, ctx = maskara.protect(prompt)
reply = llm.complete(safe)
final = maskara.restore(reply, ctx)

        

        

          

            

              
                
              
              Hugging Face
            

            
somukandula/maskara

          

          

            
Easy to plug in

            
Wrap a prompt, call your provider, restore the response. No new app architecture required.

          

          

            
Small CPU path

            
Designed for local and tiny-server demos with deterministic rules plus a compact token classifier.

          

          

            
Built for

            
Names, emails, phones, addresses, credentials, and other prompt-level sensitive spans.

            

              token classification
              local masking
              LLM middleware