“Toxic combinations” are redefining SaaS risk.
Moltbook exposure shows the pattern:
• 1.5M API tokens
• Cross-app credential leakage
• AI agents bridging trust boundaries
The real issue → permission chains across apps
Are you monitoring identity behavior at runtime?
Discuss 👇
Source: https://thehackernews.com/2026/04/cohere-ai-terrarium-sandbox-flaw.html
Follow @technadu






