Agent Security
Agent = LLM + Tools + Permissions — the next major attack surface...
Open Access
1
Threads
2
Posts
0
Today
2
Online
Moderators
ocus on Agent execution chain security: tool abuse, permission boundaries, memory contamination, task hijacking, multi‑Agent collaboration risks, and browser Agent risks.
Topic / Author
Replies
Heat
Last reply
