Tagged: Prompt injection

Articles

• Part 10Agents & Tool Use

  Building Tool-Using Agents That Work Reliably

  Tool-use reliability is a systems property: tool design, a guarded control loop, permission gates, and pass^k evals on verifiable end states.

• Part 17Security & Privacy

  Defending Tool-Using Agents Against Prompt Injection by Design

  Why model-level filters can't stop prompt injection, and how control/data separation, taint tracking and sink policies make attacks harmless.