Monk Launches Knowledge, where finance teams store the context Their AI Agents Run On

Collections teams can write down the business context an agent should apply, scope it to every customer or to one account, and see which guidance was used

Monk, the AI-native accounts receivable platform, opened Knowledge to every customer. Knowledge is where a finance team stores the context it carries about its customers and its business, so Monk’s AI agents apply that team’s judgment rather than their own.

Monk’s agents have always been configurable across playbooks, tone, cadence, when to escalate and when to stop. Knowledge holds what does not fit in a setting: the customer on non-standard terms because of a deal signed two years ago, or the account that disputes freight charges at every quarter close. That kind of detail stays with the people who know it until someone writes it down where an agent can read it.

Getting it into an agent has meant anticipating each situation in advance and writing it into a prompt. Knowledge works by retrieval instead. A team writes each piece of guidance once, decides who it applies to, and the agent finds it when the situation arises.

“Monk’s agents have always been yours to configure,” said Frank, a founding engineer at Monk. “What you couldn’t do was hand them the situational knowledge your team carries, and that is what Knowledge is. Every use is tracked, so you stay in control and always have full visibility.”

Read More on Fintech : Global FinTech Interview: AI and the future of fintech with Hugh Cumming, CTO, Vena

What Knowledge includes

Writing one piece of guidance takes about a minute. Each entry has a title, a body telling the agent how to handle the situation, and an optional line describing when it applies. From there a team can:

• apply it across the organization or to a single customer, with enable and disable controls and strict tenant and customer scoping
• pin it to a specific step in a playbook, so the right language applies to an outbound collections email or an inbound reply
• leave it unpinned, and the agent retrieves it when a customer raises something unanticipated

Monk provides templates for the situations that recur most, including payment claims, disputes, renewal disputes and missed payment commitments. Every use is recorded, so a team can see which guidance the agent applied and when.

Why it matters

Finance leaders evaluating AI for collections raise two objections: the agent will not know our policies, and we will not be able to see what it did. Both are questions about control, and Knowledge puts it with the customer. The customer writes the guidance, sets its scope, switches it off when a policy changes, and can check afterwards which entry produced a given message.

Knowledge is not model training. What a company writes is stored and read when it is relevant, it stays scoped to that company, and it can be deleted.