iam

A discussion prompt asking practitioners how they secure enterprise AI agents in 2026, specifically how to detect agents making API calls outside their intended workflow when IAM alone isn't enough, and whether teams rely on existing stacks or dedicated AI security tools.

讨论 Zenity 披露的 AWS AgentCore 安全链：暴露的 AI 被提示工程诱导查询 AWS 元数据服务并返回临时凭证，进而可访问其他 agent、密钥及写入长期记忆。作者认为核心问题不在于 SSRF 本身，而在于为何让不可信对话拥有如此大的云权限与爆炸半径。

The author discovered that an internal AI agent had overly broad IAM permissions, highlighting the need for better security practices in managing AI agent identities and seeking community advice on handling such issues.

The article highlights security concerns with AI agents in AWS environments having overly permissive access, leading to potential data exfiltration, and explores methods to control egress beyond IAM like allowlists and network-layer solutions.

This guide covers key cloud security concepts in AWS for beginners, including root vs IAM users, the Shared Responsibility model, and user permissions.

The content discusses the challenges of managing permissions and guardrails in multi-agent AI systems when agents call other agents, emphasizing the need for scalable tooling that works across mixed environments without agent rewrites.

The author argues that two distinct problems are often conflated under 'authorization for AI agents': actual access control for agents (IAM/RBAC/ABAC) and post-authorization entity-correctness (returning the wrong record despite allowed access), and asks practitioners whether this split is valid and whether existing terminology covers it.

The author discusses a common failure pattern where AI agents have valid permissions but still access or act on incorrect data or expand privileges, questioning whether current IAM/RBAC tools address this distinct problem.

This AWS article outlines five best practices for multi-tenant SaaS architecture, focusing on tenant isolation via IAM, per-tenant cost attribution, and compliance evidence to help ISVs pass enterprise procurement reviews.

A reflection on the challenge of managing permissions for AI agents in production, arguing that inventory alone is insufficient and that teams need unified control over agent actions, with ongoing interviews planned.

A discussion on how teams handle authentication and identity management for AI agents in production, covering service accounts, credential storage, approval workflows, and auditing.

A security design for AI agents accessing production cloud infrastructure using split credentials and approval gates to prevent destructive actions without human approval.