← All posts

Type · 8 pieces

Articles on AI agent authorization

Analysis and engineering notes on how AI agents get authority, how they lose it, and how task-scoped warrants bound what an agent can do.

Article

Agents are insiders. Insiders delegate.

Microsoft says AI agents should receive only the authority their task needs. A warrant chain keeps that limit intact when one agent delegates to another.

  • AI agent security
  • Delegation
Read article →
Article

Personal agents that both sides can trust

Personal agents now sign in via PACT and PAP. Task-scoped warrants add the customer's own terms, checked at the API, with receipts both sides can verify.

  • AI agent security
  • Delegation
Read article →
Article

Most companies have an AI policy. Very few can enforce it.

87% of IT leaders say an AI agent reached data it shouldn't have. Why written AI policy breaks down at runtime, and how per-task warrants enforce it.

  • Agentic governance
  • AI agent security
Read article →
Article

Let Authority Follow the Work: Tenuo for NVIDIA OpenShell

How NVIDIA OpenShell and Tenuo combine secure execution with task-scoped authority that can move across tools, sandboxes, and agent handoffs.

  • AI agent security
  • Agentic governance
Read article →
Article

Give your agent a valet key

SalesBleed leaked Agentforce CRM data without breaking an authorization check. Why AI agents need task-scoped authority, with a runnable Tenuo example.

  • AI agent security
  • Prompt injection
Read article →
Article

Per-request authorization for Temporal Nexus

Nexus allowlists decide whether a namespace may reach an endpoint. They cannot evaluate the request. Here is how task-scoped warrants close that gap without changing service contracts.

  • AI agent security
  • Delegation
Read article →
Article

Task-scoped authorization for durable AI workflows

How Tenuo and Temporal combine durable execution with task-scoped cryptographic authorization.

  • AI agent security
  • Delegation
Read article →
Article

Your AI agent is authorized to do everything wrong

Why identity-based authorization fails when agents act autonomously, and what warrant-based authorization looks like.

  • AI agent security
  • Prompt injection
Read article →