Skip to content
anythingai.space
AI Securityadvanced13 min · 9 steps

Authorization for Agents

Decide every tool call in code: may this agent do this action on this resource for this verified user — using facts the model never supplies.

close INC0047777admin rightsTHE TOOL CALL THE MODEL PRODUCEDupdate_ticket(number: INC0047777not the user'sstate: closeduser_id: E99999chosen by the model)User E12345edits own ticketsAction agentadmin service accountServiceNowINC0047777 · other ownerclosed
requesterrordatatool callsignal
1 / 9
ProblemStep 1 of 9

The agent can do what the user can't

Agents add a three-party trust problem: the user, the agent with its own permissions, and the system it acts on. The IT assistant's action agent runs on a powerful ServiceNow integration account.

A user asks it to close ticket INC0047777 — someone else's. The agent's account can, so it does. That is the confused deputy. Worse, the model filled in user_id: E99999 itself, and the tool trusted it: model-chosen identity.

Watch outAuthorization written in the system prompt (“only update the user's own tickets”) is a request, not a control.
What changed
  • +Admin integration account
  • +Tool arguments chosen by the model

Lesson 15 of 22 in AI Security