Unverified50% confidenceFactExact time
The Agent runs with the permissions of its process, not with the permissions of the user who initiated the prompt.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen an Agent needs to interface with external systems and databases, capabilities should be encapsulated into scripts to avoid repeated probing by the Agent that leads to Token consumption and slower responses.78% similarUnverifiedAn Agent compromised by prompt injection may act entirely within the authorized scope, yet execute the attacker's intent rather than the user's intent, appearing fully compliant from a permission-model perspective.77% similarUnverifiedAn agent executing a task typically requires a background process to run locally, and that process continuously holds the granted permissions rather than only during the conversation.77% similarUnverifiedAn Agent's ability to identify useful tasks does not grant it the power to allocate the permissions needed for those tasks.77% similarUnverifiedAgents possess the ability to actively execute code and access external services, requiring protection against Prompt Injection attacks and privilege escalation.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1030538API
curl https://kongchang.com/api/v1/knowledge/claims/1030538MCP
get_claim(id=1030538)