No MCP tool among 98,291 registered exposes what you'd need to rule out external-effect anomalies.

This paper builds an effect-history model separating events in the world from the runtime's observations of them, then catalogs eight recurring anomalies at the agent-tool boundary under retries, speculative execution, concurrency and partial failures (arXiv 2609.15397). Missing effects. Duplicated effects. Aborted effects that survive. Committed effects depending on provisional state later withdrawn. Measuring the standard annotation vocabulary across 98,291 tools, the fields get emitted widely but give only coarse call-level hints, and none of the required boundary capabilities is expressible. Four of the eight anomalies cannot be excluded by black-box tool invocation at all.