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

Annotations already declare readOnlyHint=false and destructiveHint=true, so the write nature is known. The description adds meaningful context beyond that: 'Never retries an upload' reassures the agent it will not re-trigger the upload, and 'private stored status' clarifies the status is stored rather than freshly queried. However it does not explain the mutation's side effects on prior status, permissions required, or the status values returned.

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, tightly packed with no filler, with the core action front-loaded and the non-retry guarantee appended as a secondary caveat.

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?

For a 3-parameter mutation tool with no output schema and annotations carrying the safety profile, the description covers the action and scope but leaves gaps: the meaning of the resulting status, what the mutation does to existing stored state, and connection_context are all unaddressed. Adequate but not complete.

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 coverage is 33% (only customer_id documented), so the description must compensate, and it partly does: 'tenant-bound opaque upload_ref' tells the agent the upload_ref is an opaque, tenant-scoped token rather than a readable identifier. It says nothing about connection_context, leaving one required-ish parameter unexplained.

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?

The description names a specific verb (refresh) plus resource (one recent direct video upload) and the effect (update its private stored status), which clearly separates it from siblings like google_ads_video_uploads_list and the upload/remove prepare/confirm pair. It is clear on its own, but does not explicitly call out the sibling tools or how it relates to the prepare/confirm flow.

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?

Usage is only implied: 'one recent direct video upload' scopes it to recent direct uploads by upload_ref, and 'Never retries an upload' hints at when it is not appropriate, but no alternative tool is named and there is no explicit when-to-use condition. It is enough to guess the intended context but not enough to confidently route between it and google_ads_video_uploads_list.

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