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

Annotations already declare idempotentHint=true and non-destructive, but the description adds material context beyond them: this call actually spends credits and enqueues an asynchronous job, and the client must persist the idempotency key beforehand. That credit-spend warning is the highest-value behavioral disclosure for a mutation tool. It omits failure behavior (what happens on stale expected_credits or pricing_revision_id mismatch) and how to observe the enqueued job.

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, front-loaded sentences with no filler: the side effect first, then the operational instruction. The wording is tight, though "enhancement" is too abstract to be self-explanatory and costs a little clarity for the brevity gained.

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?

This is a credit-spending, asynchronous creation call with an eleven-field nested config, no parameter descriptions, and no output schema, so the description should explain the generation parameters and what the caller receives (a generation id to poll via get_generation). Neither is covered, leaving the agent able to identify the tool but not to fill the payload confidently.

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 0%, so the description carries the full burden, yet it only addresses idempotency_key ("persist the key before calling"). The nested generation object's eleven fields — expected_credits, pricing_revision_id, resolution, fps, source_result_id, and the smoothing flags — get no explanation of meaning, units, or constraints, which is a serious gap for the two required fields that govern pricing and idempotency.

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?

"Spend credits and enqueue enhancement" gives a concrete verb and effect, and the phrase "after user authorization of the estimate" ties it to the estimate step, distinguishing it from estimate_generation. It stops short of saying what is being generated (video/upscale) or what object is returned, so the resource is only half-specified.

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 precondition (user must have authorized an estimate) is stated and implies this is the step following estimate_generation, which is useful ordering guidance. However, it never names the estimate tool as the prerequisite explicitly, nor states when not to call it (e.g., insufficient credits, missing pricing revision), leaving the agent to infer the workflow.

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