Wisemail

Inspiration

Compliance failures cost financial institutions billions every year, often due to simple communication mistakes in emails. Existing tools detect issues only after messages are sent, which is too late.

We wanted to explore how AI could help prevent these mistakes in real time instead of reacting after the fact.

What it does

Wisemail is an AI assistant that detects and fixes non-compliant emails in real time—before they are sent.

It highlights risky sentences, explains why they may be non-compliant, and suggests safer alternatives.

How we built it

We built Wisemail as a Chrome extension.

• A content script interacts with text inside email clients
  
• User input is sent to Claude for analysis
  
• The model identifies potential compliance issues and suggests improvements
  
• A simple interface displays feedback inline
  

Challenges we ran into

• Accessing and analyzing live text input inside web email clients
  
• Making responses fast enough to feel real-time
  
• Keeping suggestions clear and useful without overcomplicating them
  

Accomplishments that we're proud of

• Successfully integrating AI into a live writing workflow
  
• Building a working prototype that provides instant feedback
  
• Keeping the experience simple and non-intrusive
  

What we learned

• Compliance is often about small wording details
  
• Real-time feedback is much more impactful than post-review
  
• Simplicity is key when designing tools people use while writing
  

What's next for Wisemail

• Expand to more compliance domains
  
• Refine the user experience and interface
  

Built With

• claude