← All 186 courses
SEC-502 Security

Securing Agentic AI Systems

Agents act — so their failures act too. Engineer bounded autonomy: least-privilege tool access, action budgets and human approval gates, sandboxed execution, audit-grade action logging, and kill-switch design — stress-tested by driving a deliberately over-trusted agent into contained failure and fixing it.

B5Advanced
5modules
35skill atoms
2role journeys
Curriculum

What this course covers.

5 modules, 35 named skill atoms. Expand any module to see them.

1Agent Risk Model7 skill atoms
action surfacecascading-failure scenariostool inventory and reachable side effectsreversible vs irreversible action classificationdelegation & multi-agent trust riskpersistent memory poisoningrunaway loop and cost failure modes
2Least-Privilege Tools7 skill atoms
scoped credentials per tooldry-run modesper-tool identity instead of one shared principalread and write tool separationparameter allow-lists & path constraintsshort-lived tokens per invocationruntime capability revocation
3Autonomy Boundaries7 skill atoms
budgetsapproval gatesrefusal rulesstep, time & spend budgets per taskrisk-tiered human-in-the-loop gatesescalation instead of guessingidempotency & retry limits
4Execution Sandboxing7 skill atoms
isolation layersegress controlmicroVM or container isolation per sessionno-network default with allow-listed egressephemeral scoped workspacesresource quotas against exhaustionsecrets injected outside model context
5Audit & Kill Switch7 skill atoms
action ledgersanomaly haltsrecovery drilltamper-evident append-only action logcorrelation ids across prompt, tool call & resultglobal and per-agent disable switchrollback & compensating actions
Where it fits

SEC-502 in the role journeys.

This course appears in 2 of our 45 role journeys. Here is what a learner takes immediately before and after it in each.

Agentic AI Engineer

Capstone stage
AI-509SEC-502AI-510

AI Security Engineer

Professional stage
SEC-303SEC-502AI-404

Roles this course serves

The capability ladder

This course is authored to band B5.

Every course we run is written to one rung of the CASI ladder, so a plan can be assembled to take a team from where they are to where they need to be.

What do B1–B6 mean?The CASI Capability Ladder — click to expand

Every course targets a band on the CASI Capability Ladder — our six-band proficiency scale, anchored to open standards (O*NET, ESCO, NICE, NIST AI RMF, Bloom's). A band tells you how deep a course goes, and what evidence proves it.

What the learner can doTypical evidence
B1
AwareUnderstands concepts and vocabulary; uses tools with guidance
Knowledge checks
B2
FoundationPerforms standard tasks correctly in familiar contexts
Guided labs, autograded exercises
B3
PractitionerDelivers complete pieces of work independently
Scenario labs, proctored hands-on exams
B4
ProfessionalHandles production-grade complexity, trade-offs and failure modes
Break-fix drills, design defenses
B5
AdvancedEngineers systems end-to-end under constraints; leads others
Rubric-scored capstones, vivas
B6
ExpertSets direction; recognised authority across teams
Portfolio + panel evaluation

A note on B6. Courses in this catalog target B1–B5. B6 is not taught — it is recognised, through a portfolio and a panel, once someone is setting direction for others. Every journey here is built to land a learner at B5.

Next step

Run SEC-502 for your team.

This course runs at several lengths depending on how deep you need to go and how much of it your people already have. Tell us who is being trained and we will scope it.

Add it to a training plan Talk to our team Check your team’s level free