Request Manager Agent

Fulfilling a privacy request, whether deleting someone's records or pulling everything a company knows about them, requires reaching the systems where that data actually lives. For many companies that data sits in databases inside their own private network, and it is the most sensitive data they hold. This agent runs inside that network and does the work in place, executing the access, deletion, and lookup operations a request needs directly against the customer's own databases.

That placement is the whole point. The platform can orchestrate privacy requests against data it never sees, copies, or stores centrally; the customer keeps full custody of their own records, and only the results of an operation travel back. To make this work even in the most locked-down environments, the agent talks to the platform over outbound connections only, so it runs inside a private network without opening a single inbound port. A security team that would never expose its databases to a vendor can still adopt it.

Making remote, in-network execution reliable is where the real engineering sits: each unit of work is leased so two agents never process the same request twice, and every result is correlated back to the exact operation that asked for it. The agent reuses the same library of database connectors used across the platform, fed through an adapter that supplies the credentials and queries for each request, so the code that actually talks to customer systems is shared rather than reimplemented.