New cohorts enrolling monthly — onsite, online & hybrid formats. Modern SQA Formula Bootcamp 2026 Registration ongoing Corporate teams: ask about custom bootcamp pricing.
AI quality engineer tracing an agent plan tool calls permissions and approval gates
SQA, Test Automation & SDET · Course #059

Agentic AI Testing

Effective practitioners need to know what to do, why it matters and how to recognise a weak result. Agentic AI Testing develops that judgement by helping learners evaluate AI agents that plan, use tools, retain state and take actions across multi-step workflows. The programme integrates agentic systems, intended goals, success criteria and prohibited actions and scenario and trajectory evaluation for variable paths rather than teaching them as isolated topics. Participants map goals and permissions, test planning and tool selection, simulate tool failures, inspect traces, probe excessive agency and verify approval, rollback and stop controls. They create scenario-based evidence for both task success and safe action boundaries in systems whose path may vary between runs.

Enroll Interest Back to Courses

Overview

Effective practitioners need to know what to do, why it matters and how to recognise a weak result. Agentic AI Testing develops that judgement by helping learners evaluate AI agents that plan, use tools, retain state and take actions across multi-step workflows. The programme integrates agentic systems, intended goals, success criteria and prohibited actions and scenario and trajectory evaluation for variable paths rather than teaching them as isolated topics. Participants map goals and permissions, test planning and tool selection, simulate tool failures, inspect traces, probe excessive agency and verify approval, rollback and stop controls. They create scenario-based evidence for both task success and safe action boundaries in systems whose path may vary between runs.

Course Highlights

Agent goal and path evaluation

Tool selection and parameter checks

Memory and state testing

Permission and excessive-agency controls

Failure recovery and rollback

Modules & Curriculum

Agentic systems: plans, tools, state and autonomy

Intended goals, success criteria and prohibited actions

Scenario and trajectory evaluation for variable paths

Tool selection, arguments and result interpretation

Memory, context persistence and cross-session risks

Permissions, identity and least-agency design

Tool failure, timeout, retry and recovery behaviour

Human approval, interruption and rollback controls; trace analysis, reproducibility and regression sets

Red-team scenario and agent-safety evidence

Learning Outcomes

  • Model an agent's goals, tools, state and action boundaries.
  • Design evaluations for planning quality and task completion.
  • Verify tool selection, parameters and downstream effects.
  • Test memory accuracy, leakage and state transitions.
  • Probe excessive agency, privilege and unsafe action chains.

Ready to Build Your Future-Ready Skills?

Test agent goals, tool actions and safety controls across variable multi-step workflows.

Enroll Interest

Related Courses

Beginner Junior QA learners mapping assurance activities across a software delivery lifecycle

Software Quality Assurance (SQA) Fundamentals

View Course
Beginner Software team tracing requirements through development testing release and maintenance stages

Software Development Life Cycle (SDLC)

View Course
Beginner QA trainees organising test planning execution defect tracking and closure artefacts

Software Testing Life Cycle (STLC)

View Course