Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare destructive=true, openWorld=true, idempotent=true, readOnly=false; the description adds value the annotations cannot convey — single-shot semantics and the exact failure-recovery protocol (recover the original response rather than re-issuing a replacement). The retry prohibition sits in mild tension with idempotentHint=true, though 'replacement refresh' (new request) is not identical to a safe retry, so it is not a contradiction.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Two tight sentences with no filler, and the critical non-retry rule is placed where it will be read. The opening sentence largely restates the title plus 'once', so a small amount of the text is redundant.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema and rich annotations cover results and safety, so the description needn't explain returns. But for a destructive, non-retryable operation the description omits the prerequisite flow — obtaining the connectionId/operationId and catalog version from a prior request — which an agent needs to call it correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With six parameters (five required, nested request object with action/connectionId/operationId) at roughly 50% schema description coverage, the description names no parameter at all and does not explain where connectionId/operationId or expected_catalog_version come from. The required request_key and work_id are entirely unexplained in prose.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Refresh the selected analytics authorization') plus scope ('once'), which separates it from broader siblings like genads_analytics_request and genads_import_analytics. It never names a sibling explicitly, so differentiation is implied rather than enforced.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives a concrete decision rule for the ambiguous case: if acceptance is unknown, recover the original response or reconnect, and never submit a replacement refresh as a retry. It does not say which tool implements recovery (native.genads_recover_analytics exists) or when to call this versus genads_analytics_connections, so routing to the actual alternative is left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.