A Content Agent operates as an always-on monitoring and drafting layer. It watches designated external sources — regulatory bodies, standard setters, and research organisations — and detects changes that affect CA ANZ's content. When a change is detected, it generates a structured brief and routes it to the appropriate content owner. It can also draft first-pass content — storyboards, module outlines, and update summaries — for human author review. The Content Agent never publishes. It informs and accelerates; humans write and approve.

An Assessment Agent operates across two distinct modes. In Mode A (Knowledge Check), it engages candidates before a formal assessment — presenting questions mapped to learning outcomes, analysing responses, and delivering an instant readiness report with a personalised study path. No PII is required and no human gate is needed. In Mode B (Marker Assist), it sits inside the formal marking pipeline — generating structured draft feedback on candidate submissions for human marker review and approval before release.

A Coach Agent engages candidates or members in structured conversational experiences — scenario-based roleplay, ethical dilemmas, stakeholder simulations, and reflective coaching dialogues. It plays a role (client, regulator, sceptical board member, non-finance stakeholder) and responds dynamically to the learner's choices. After each interaction, it provides a personalised debrief anchored in the CA Capability Model. The scenario library, personas, and debrief frameworks are designed and approved by the teaching or content team before deployment.

An Insights Agent analyses qualitative and quantitative data — assessment performance, learner reflections, market signals, and SME interview transcripts — and surfaces patterns that would otherwise require significant manual effort to identify. It clusters, synthesises, and reports. It does not make decisions; it makes data legible so that humans can make better decisions faster. In the CA ANZ context, Insights Agents serve the teaching team (cohort performance patterns), the product team (emerging CPD demand signals), and the content team (SME knowledge extraction).

A Mapper Agent is a connective intelligence layer. It takes what is known about a learner — self-assessment results, subject completions, exam performance, CPD history, career stage, or employer capability framework — and maps it against a structured model: the CA Capability Model, subject learning outcomes, CPD catalogue, or a B2B client's internal framework. It then generates a personalised output: a study plan, a CPD pathway, a readiness assessment, or a capability gap report. The Mapper Agent makes the invisible visible — showing learners and employers where they are and what comes next.