Thirdeye AI pay-per-call inference

MCP guide

An agent can be given a model context protocol server. The tools it exposes are treated as untrusted input.

What is stored

A list of server entries on the agent row. Each entry is a name and a URL; there is no separate credential store for a server, so a server that needs a token is out of scope today.

Guards

ControlBehaviour
SchemeOnly https. Anything else is refused before a socket is opened.
DestinationPrivate, loopback, link-local and metadata ranges are refused, and the check runs again after every redirect.
Reply sizeCapped, so a hostile server cannot exhaust memory.
SpendA paid tool call is bounded by the agent's per-call cap and its allowlist. An unlisted host cannot be paid.
StepsThe run has a step cap and a budget cap. A loop stops rather than running until the credit is gone.

The same destination policy applies to the generic fetch tool, so a URL that appears in a prompt cannot reach an internal service.