generate_video

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

The description adds genuinely useful context beyond the annotations: this call spends credits (a side effect with real cost) and that requestId should be reused on retries, which gives practical meaning to `idempotentHint: true`. It does not disclose that generation is likely asynchronous or what the call returns, and the presence of `get_task`/`list_tasks` siblings suggests the agent needs a polling step the description never mentions.

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, front-loaded with the cost-bearing action and followed by the operational caveat. Nothing is redundant and no sentence is filler.

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 two-parameter call whose `input` object is a complex three-way discriminated union with 0% field documentation and no output schema, the description under-delivers on what happens after invocation and how the variants differ. The credit-cost and requestId guidance are the right instincts but leave the async/return behavior and variant selection unresolved.

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 must carry meaning for `input` and `requestId`. It does explain the role of `requestId` as a retry-safe identifier and sketches the three input modes, but it says nothing about required fields (`prompt`, `modelId`), resolution/duration/aspect-ratio choices, or the distinction between the first-frame and reference variants.

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 concrete verb and resource ('Spend website credits on a ... video') and enumerates the three supported modes (text, first-frame, reference), which map cleanly onto the three `anyOf` variants in the schema. It is clearly a generation tool with a cost, distinguishable from `estimate_generation` and `get_task`, though the sibling differentiation is only implicit rather than stated.

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?

'Confirm the estimate' is a workflow hint that implies running `estimate_generation` first, and 'reuse requestId on retries' gives retry guidance. However, there is no explicit statement of when to use this versus the estimate or task-polling siblings, and no exclusions or prerequisites beyond those fragments.

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