> For the complete documentation index, see [llms.txt](https://docs.scaleconnect.com.au/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.scaleconnect.com.au/scaleconnect-docs/scaleai/ai-assistant.md). # AI Assistant Every client in your book gets their own AI assistant, powered by an institutional-grade audited and certified model with knowledge of their properties, loans, personal finances, cashflows, goals, meeting history, and every review you've sent them. Ask it anything — from portfolio-level equity positions to month-by-month budgeting — and get answers grounded in that client's real numbers. ## **How it Works** 1. **Queries relevant client data across 50+ tools** — strategy inputs, portfolio positions, cashflows, loan structures, goal progress, timeline projections, and your broker settings — fetched in parallel in a single query 2. **Traces causality through a 355-node dependency graph** — identifies which events, inputs, and external factors are driving a metric, then maps every upstream cause and downstream impact across cashflow, equity, tax, and borrowing power 3. **Passes everything through a five-layer knowledge system** — 967 field definitions, interpretation logic, Australian property and personal finance principles, and ScaleAcademy educational content — so it reasons through the data, not just retrieves it 4. **Delivers a cohesive response in the context of the client's goals** — every answer is shaped by their strategy, their timeline, their targets, and the history you've built up through notes, transcripts, and past reviews 5. **Liaises with a fleet of specialised AI agents** — from automated reviews and meeting scheduling to proactive workflow triggers and your [Broker Brain](/scaleconnect-docs/scaleai/broker-brain.md) — the assistant coordinates across the full ScaleAI system on your behalf The result is an answer grounded in that client's actual data, shaped by the context you've provided, and reasoned through an institutional-grade [audited and certified](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/scaleai/data-integrity.md) financial model — not a generic response. *** ## The AI Client Intelligence Loop Your AI Assistant doesn't work in isolation — it's the centre of a feedback loop that makes every client interaction smarter. 1. **You ask questions** → the AI draws on the client's portfolio data, notes, and past reviews to respond 2. **You save responses** → the **Add to Notes** button stores insights into the client's [knowledge base](/scaleconnect-docs/scaleai/ai-assistant/client-notes.md) 3. **Meetings happen** → the [AI Note-Taker](/scaleconnect-docs/scaleai/ai-assistant/ai-note-taker.md) captures transcripts and extracts action items 4. **Reviews generate** → [AI Reviews](/scaleconnect-docs/scaleai/ai-assistant/ai-reviews.md) reference past notes and conversations to produce personalised analysis 5. **Scheduling triggers** → the [AI Scheduling Bot](/scaleconnect-docs/scaleai/ai-assistant/calendar-agent.md) monitors readiness signals and books follow-ups automatically Each step feeds context back into the assistant, so the next conversation starts with a richer picture than the last. *** ## What It Has Access To The AI draws from both sides of the platform — property portfolio analysis and personal financial budgeting: * **Portfolio** — properties, current values, purchase history, ownership structures * **Loans** — balances, rates, lender, type (P\&I / IO), fixed or variable, tenor, commission * **Personal finances** — income, expenses, savings, dependents, budgeting, debt obligations, sole trader income, HECS * **Cashflows** — personal, property, and portfolio-level projections across financial years * **Goals** — property targets, savings milestones, lifestyle goals, timeline, completion status * **Strategies** — every scenario configured in the simulator, with full timeline data * **Notes and transcripts** — everything in the client's [knowledge base](/scaleconnect-docs/scaleai/ai-assistant/client-notes.md), including meeting transcripts, manual notes, and saved AI summaries * **Past reviews** — every [AI Review](/scaleconnect-docs/scaleai/ai-assistant/ai-reviews.md) you've generated and sent for this client *** ## Your Client List The AI Assistant monitors every client's position and orders your client list by readiness. Clients approaching key thresholds, with expiring fixed rates, or due for a review surface to the top — you don't scan your book manually, the AI prioritises it for you. Select a client to open their profile and switch to the **AI Assistant** tab to start a conversation. {% @arcade/embed url="" flowId="TlNbsxJT73XaeNeFPhDG" %} *** ## Using the Chat ### Strategy selector Every conversation is scoped to a strategy. The **strategy selector** sits below the input box — switch between the client's scenarios to focus the AI's answers on a specific strategy. Comparing two approaches is as simple as switching the selector and asking the same question twice. {% @arcade/embed url="" flowId="Wo8LYXed2T30CRMZcogz" %} ### Suggested prompts When you open a client's AI Assistant, you'll see suggested prompts tailored to that client's current position. These are generated based on what's relevant right now — approaching thresholds, recent changes, upcoming milestones — so you can get straight to the most useful question without thinking about what to ask. {% @arcade/embed url="" flowId="WAZ5z2ChngQyE7Gj7DRO" %} ### Voice Prompting The AI Assistant supports voice input — click the microphone button to speak your question instead of typing. Useful during meetings when you want a quick answer without breaking eye contact. {% @arcade/embed url="" flowId="droSbqHolVu24IKrimgs" %} ### Response Actions Each AI response includes action buttons:
ActionWhat It Does
CopyCopies the response to your clipboard
ShareShares via your device's share sheet
RegenerateRe-sends the question for a fresh response
Add to NotesSaves the response as a note on the client's file
{% @arcade/embed url="" flowId="krv4gNOD1OTSnX7ZEHNI" %} ### Chat History Each client maintains their own conversation history in the left sidebar. Conversations are grouped by date — pick up where you left off, start a new thread, or revisit a past deep dive. Switching between clients keeps each history separate, so context never bleeds across. {% @arcade/embed url="" flowId="5SeWwFqfSeqJmoYfhwAm" %} *** ## Suggested Responses in Client Chat When a client messages you through the [Client Chatbot](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/scaleai/product-docs/scaleconnect-docs/scaleai/client-chatbot.md), the AI Assistant generates suggested replies based on the client's portfolio data, recent activity, and conversation context. You can send a suggestion as-is, edit it, or dismiss it and write your own. This means the same intelligence that powers deep analysis also helps you respond to day-to-day client messages faster — without sounding like a bot. {% @arcade/embed url="" flowId="W08C2YV40MH30Lbnm0qt" %} *** ## What You Can Ask ### Portfolio and property * "Give me a snapshot of this client's portfolio" * "How has the investment property in Brisbane changed in value since purchase?" * "What's the current LVR across all their properties?" * "Break down each property's cashflow contribution" ### Personal finances and goals * "What does their personal cashflow look like this financial year?" * "Are they on track for their deposit target?" * "How are their living expenses tracking against income?" * "What's the impact of their HECS repayments on borrowing power?" ### Strategy deep dives * "Compare the base strategy with the refinance scenario" * "What's driving their negative gearing position?" * "Walk me through how their equity position changes over the next five years" * "What are the key risks in this strategy?" ### Review and meeting prep * "Generate a review summary for this client" * "What's changed since the last review?" * "What should I focus on in our next meeting?" * "Summarise the action items from our last conversation" {% hint style="info" %} The AI always leads with the client's actual numbers before explaining a concept. You won't get a textbook answer without context. {% endhint %} *** ## Common Use Cases
ScenarioHow Brokers Use It
Before a meetingPull up the client's AI Assistant, ask for a portfolio snapshot and recent changes. Walk into the meeting already briefed.
During a meetingAsk live questions — "What's their borrowing power?" or "How does refinancing at 5.5% change things?" — without leaving the conversation.
After a meetingThe AI Note-Taker captures the transcript. Ask the assistant to summarise key action items or save a response as a note.
Between reviewsCheck in on a client's position without generating a full review. "Has anything changed since last quarter?"
Explaining to a clientGet a plain-language breakdown of a metric or concept with the client's own numbers, then share it via the Add to Notes action.
--- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.scaleconnect.com.au/scaleconnect-docs/scaleai/ai-assistant.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.