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

Annotations declare readOnlyHint=false and destructiveHint=false, but the description adds substantive context the annotations do not: the operation is a *permanent* removal and it is gated behind a confirm step, so this call is a safe staging action. That resolves the tension an agent would feel between 'remove' and destructiveHint=false. It still omits what prepare actually produces (a token/plan for confirm).

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 short sentences with zero filler, and the non-destructive nature of the call is front-loaded right after the purpose. Nothing redundant or padded.

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?

There is no output schema and no annotation coverage of the return value, so the description should ideally say what prepare yields for the confirm step. It covers the safety model well but leaves the workflow hand-off and the two undocumented parameters unexplained, which is the minimum-viable level for a 3-parameter mutation-prep tool.

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?

Schema description coverage is only 33%: only customer_id is documented in the schema, while upload_ref and connection_context carry no description. The phrase 'one direct video upload' loosely hints that upload_ref identifies a single upload, but no format, sourcing, or constraints are given, and connection_context is fully unexplained in both schema and description.

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 ('Prepare permanent removal') and resource ('one direct video upload'), and the phrase 'Does not remove until confirm' distinguishes it from the sibling confirm tool. An agent can tell this apart from google_ads_video_upload_remove_confirm and google_ads_video_uploads_list without opening any schema.

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?

The sentence 'Does not remove until confirm' implies the two-step prepare/confirm workflow, which is genuine routing guidance. However, it never names the confirm tool explicitly, states prerequisites (permissions on the customer), or says when this is preferable to other removal paths. Usage is implied rather than stated.

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