Solution introduction
Agents that do advisor work — preparing meetings, writing up notes, filling flows, fetching numbers — inside guardrails written in code rather than in a prompt, with a readable record of every run
Most of an advisor's week is not advice. It is preparing for meetings, writing up what was said afterwards, filling in flows, fetching numbers from other systems and keeping the client record straight. Quantfolio Platform's agents take that work on
Guardrails in code, not in a prompt
In most AI tools for advisors the rules are a skill or a prompt: text the model is asked to follow, and usually does. In a regulated industry, usually is not good enough. In the platform an agent works through the same flows and the same checks as an advisor. A portfolio outside the client's risk bounds is stopped by the platform, not by the model's judgement. A tool the agent has not been given does not exist for it. An instruction can be misread or talked around; a check in code runs the same way every time.
Two built-in agents
The Global Agent works across all clients, with tools for an overview of the organisation's activity, statistics, prepping the advisor's week and prospecting. The Client Agent works one client at a time, with tools for filling flows, logging activities, creating portfolios, prepping meetings and analysing existing positions. Both start from the client's own history — notes, transcripts, documents, the whole file — so they rarely ask for what you already hold.
Set them in a system
Predefined prompts for recurring tasks, scheduled tasks that run at a given time and frequency, and event-based tasks that fire when something happens. Ad hoc prompting stays available alongside them.
A record you can read
Every piece of work an agent does leaves the request as it was asked, every tool it called in order with whether each succeeded, the result exactly as the advisor saw it,
and the run itself — the model, the tokens, the time and the cost. Not a summary the agent wrote about itself. The conversation stays on the client and is shared with the team, so a colleague picking the client up, or compliance checking the work, can follow what was done and how.
Your models, or ours
The platform is AI-agnostic. Use the models that ship with it — all routed through the EU, running on two EU clouds with automatic failover — or connect your own provider with your own keys through Azure OpenAI, OpenRouter or any OpenAI-compatible API. You set the default model for agent work and a lighter one for translations and summaries; models you do not want offered can be switched off.
Your agents, our platform
A built-in MCP server lets your own AI agents operate Quantfolio Platform inside the same guardrails that apply to everyone else. Connect MCP servers from your own ecosystem and the platform's agents can use their tools while they work.
