Easier to read with cream paper, a tint, more contrast or larger text? Choose your colours and text, once, for every page.

Free module

Securing agents

Learn AI agent security hands on: threat models, data exfiltration, confused deputies, untrusted tools, privilege separation and red teaming.

7 lessons475 min lessonSoftware engineering

The lessons

Each lesson is free to read in full. Create a free account to run its commands in a live lab and to sit its quizzes and tasks.

  1. 1 · Container lab

    Threat modelling an AI agent

    I can map where an agent can be attacked.

    Read the lesson

  2. 2 · Container lab

    Preventing data exfiltration from AI agents

    I can close the routes data leaves an agent by, using an allow list of destinations that code checks.

    Read the lesson

  3. 3 · Container lab

    Confused deputy problem in AI agents

    I can stop an agent acting with more authority than the person asking, by checking each action against the asker.

    Read the lesson

  4. 4 · Container lab

    Securing untrusted tools for AI agents

    I can decide whether to trust a tool I did not write, by pinning its definition and reviewing what it says.

    Read the lesson

  5. 5 · Container lab

    Privilege separation for AI agents

    I can split an agent so untrusted text never meets the power to act.

    Read the lesson

  6. 6 · Container lab

    Red teaming an AI agent

    I can measure how often an attack works, and whether a fix helped, over a recorded set of attacks.

    Read the lesson

  7. 7 · Container lab

    Measuring AI agent defences

    I can score several agent defences on a labelled set, counting caught attacks, missed attacks and false positives.

    Read the lesson

Worked in a real lab

Every lesson of this module is worked in a lab started for you alone. The labs it uses:

Teaching a class? Set this module for your students with a teacher guide, marked assignments and a lab for every student.