Superagent

Headline: Red teaming for AI agents

Target: Companies deploying AI agents in production

Category: AI agent security and testing

Problem: AI agents in production can have data leaks, harmful outputs, and unwanted actions that need to be identified and fixed before users encounter them.