Skip to content

Who it's for

The demo and these docs serve four audiences. Each audience cares about a different surface. The links below jump straight into the part of the documentation built for you.

If you sell solar PV, batteries, EV chargepoints, or any other home-improvement product where the customer typically finances a high-value install, this is for you.

“Retailer” here means the firm whose installer is on the customer’s doorstep. You introduce the customer to the broker; you don’t hold the FCA permission yourself. The agent moves the journey from a clunky web form to a doorstep conversation that completes on the customer’s phone before the installer has packed up.

Start with Adoption path for retailers. Then Integration patterns and the Pilot playbook.

If you hold the FCA permission and arrange consumer credit through a panel of lenders for home-improvement / renewables introductions, this is for you. You’re the regulated entity. The retailer introduces the customer; you arrange the finance.

The agent is a stronger Consumer Duty surface than a form: same disclosures every time, same vulnerability check every time, replayable for compliance evidence. The trust gradient lets regulated moments render inline (high-trust user agents) or on the broker’s own surface (low-trust user agents) without changing the underlying journey.

Start with Adoption path for brokers. Then Lender panel integration, Disclosure publishing, and Audit & replay integration.

For the regulatory framing, Consumer Duty and CONC are the load-bearing chapters.

If you sit on a broker’s panel (prime, near-prime, or sub-prime) and you’re wondering what changes when the broker swaps a form-based journey for an agent-mediated one, this is for you.

Less changes than you might think. You already receive structured applications via your existing broker integrations. The agent doesn’t change the shape of what you receive. It changes what the broker can prove about how it was collected, and it changes the surface area you can offer counter-pricing on.

Start with Waterfall protocol. Then Counter-offer mechanics and Decision API adapter.

If you’re reviewing whether AI-mediated agentic journeys meet the regulatory bar, the Regulatory section is the entry point. Then Safety and Privacy for the cross-cutting concerns.

The strongest concrete artefact is the replay engine. Section 7.7 of the agentic credit broking protocol calls for replayability; the demo implements it. Open any completed case at https://lending-agent.vercel.app/audit/[sessionId] and click “Run statistical replay”.

Curious technologist? Architecture overview is where the interesting design decisions live. Want to deploy your own copy? Deploy guide.

If you’re thinking about a similar problem in a different vertical (mortgage broking, motor finance, insurance arrangement), the same pattern applies. Get in touch.