| Framework Home | Research Index | Cross-Pillar Governance | AISM | NEXUS | Dashboard |
Current framework context: AI SAFE² v3.1. This research note preserves its original publication date, evidence, and historical framework references. Use current v3.1 normative control and profile documents for implementation or conformance decisions.
ID: RN-2025-007 | Related Control: [P4.T1.2_ADV] | Status: Verified
Over-Privileged Agents: Developers often grant agents “Admin” access to APIs (e.g., GitHub, Jira, AWS) because it is convenient. If the agent is hijacked, the attacker inherits full admin rights.
We apply Zero Trust Principles to Agent permissions.
Agents default to “Read-Only.” If an agent needs to perform a “Write” action (e.g., Deploy Code), it must request a Temporary Token.
The request is checked against a “Behavioral Baseline.” If a Customer Support agent requests Database Write access, the system auto-rejects it based on role mismatch.
| Previous research note | Research Index | Next research note |
| Framework Home | Cross-Pillar Governance | NEXUS | Challenge Lab |
AI SAFE² v3.1 | Cyber Strategy Institute