FlatClaw, Private AI Platform
All use case spotlights
Knowledge SearchCompliance & ApprovalsFinancial Services

Community financial institution

≈ $2.4B in assets (approx.) · 30 branches · ~400 employees

Governed core-banking assistant

Staff ask questions of the core banking system through governed MCP tools. Anything consequential is composed and paused for a human, so the assistant is useful without being able to act alone.

Organization
Community financial institution
System
Core banking platform with a well-defined request surface
Runs on
The institution's own cloud tenancy
Scope
Governed assistant with approval-gated actions
The situation

Where they started.

Staff wanted to ask the core system questions in plain language: account status, transaction history, exception queues. The institution's IT security team, reasonably, would not allow an assistant that could act on the core system on its own, and would not allow the core's data to leave the tenancy.

What FlatClaw does

What was built.

  • MCP connectors that mirror the core platform's real request shapes rather than inventing an abstraction over them.
  • Read paths open to the roles that need them, with per-user credentials so an answer never exceeds what that person could see themselves.
  • Write paths composed by the agent and paused in the approval queue; on approve, the request replays with the approver's own credentials.
  • Every decision and every replayed request recorded in the audit log.
Results

What changed.

  • A useful assistant that cannot act alone.
  • Security review passed on the strength of the approval engine and the audit trail.
  • Consistent answers for staff who previously escalated to a specialist.
  • A pattern the institution can extend to other systems.
Why private

Account data does not leave a bank's boundary, full stop. The agent's usefulness had to be earned without any of that data reaching a third party.

Runs on

The institution's own cloud tenancy

The stack

Components involved.

Core-banking MCP connectorsApproval enginePer-user credential vaultFlatClaw PortalPrivate inference on a dedicated GPU
Your workflow

Have one like it?

Every spotlight started as a conversation about a process nobody liked doing, under a data-locality constraint.