📄 中文摘要
随着 AI 代理的自主性增强，确保其操作的可追溯性变得愈发重要。共享服务令牌在演示中可能有效，但在生产环境中却会破坏问责机制。当多个代理使用相同凭证访问 GitHub、Jira、Slack 和内部 API 时，基本的安全问题难以回答，例如：哪个代理批准了此操作？它是在谁的授权下行动的？为了解决这些问题，提出了结合密码身份和基于角色的访问控制（RBAC）的方法，以确保每个代理的操作都有明确的责任和审计轨迹，从而增强系统的安全性和透明度。
用于主权 AI 代理问责的密码身份与 RBAC
出处: Cryptographic Identity & RBAC for Sovereign AI Agent Accountability
发布: 2026年3月28日
📄 English Summary
Cryptographic Identity & RBAC for Sovereign AI Agent Accountability
As AI agents become more autonomous, ensuring accountability for their actions is increasingly critical. While a shared service token may suffice for demonstrations, it undermines accountability in production environments. When multiple agents use the same credential to access platforms like GitHub, Jira, Slack, and internal APIs, fundamental security questions become difficult to answer, such as: Which agent approved this action? Under whose authority did it act? To address these challenges, a method combining cryptographic identity and Role-Based Access Control (RBAC) is proposed to ensure that every action taken by an agent has a clear responsibility and audit trail, thereby enhancing the security and transparency of the system.