Use the rubric before export to identify weak reasoning and unsupported conclusions.
PORTFOLIO PRACTITIONER LABSecureGenAI
SecureGenAI
GenAI, RAG and agent security
Review an enterprise RAG assistant and tool-using agent for retrieval, hallucination, leakage, prompt injection, permissions and unintended actions.
Start SecureGenAI 60–75 min
Weeks 10–11 · Fictional scenario · reviewer challengeYour answers save in this browser. Use no real personal, client or confidential data.
YOUR ROLE
You are the GenAI security and governance reviewer.
Threat-model the RAG and agent architecture, define security controls and decide whether the pilot is safe to release.
- RAG Risk Checklist
- Agentic AI Control Checklist
- System Card
- AI Testing and Evaluation Plan
Training simulation only.All organisations and system details are fictional. This is not legal advice or a compliance guarantee.
PRACTICE MODELoading
Restoring your saved mode and lab work…
01
RAG and LLM risk
RAG Risk Checklist
02
Agentic AI
Agentic AI Control Checklist
03
Prompt injection and security
System Card
PRACTITIONER STANDARD
Build it. Challenge it. Defend it.
This lab includes practice setup, professional review criteria, challenge questions, traceability and readiness evidence.
Can the RAG or agent system resist untrusted content and prevent unauthorised actions?
Evidence must be relevant, attributable, current and strong enough to support the decision.