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

Goes well past the annotations: async fire-and-return semantics, token cost driven by server-measured clip length, uncharged refusal for unmeasurable files (WebM/unreadable header) with the re-encode remedy, one media-ingest slot for external URLs, and the idempotency_key replay rule. Nothing contradicts readOnlyHint=false / destructiveHint=true / idempotentHint=false; the replay caveat is nuance the annotations cannot express. Slightly held back only because the interaction between destructiveHint and token spend is left implicit.

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?

Front-loaded with purpose and WHEN before the workflow and cost rules, and the capitalised labels make it scannable. It is dense and partly duplicative, though: the COST block already says the call spends the user's tokens, which the separate SPENDS TOKENS block repeats.

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 an async, token-charging, open-world mutation with no output schema, it supplies the missing return contract (job id now, url via card otherwise), the follow-up call, the cost gate and the idempotency contract. An agent has everything needed to invoke and sequence 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?

Schema coverage is 100%, so the baseline is 3; the description still adds operational meaning the schema lacks: price first with estimate_only and ask in the same message, and never reuse an idempotency_key for a new generation. The upscale_factor ceiling (7680px) is stated in both, which is redundant but not harmful.

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?

Opens with a specific verb and resource plus measurable scope: increases an existing video's resolution to 2x/3x/4x and returns the job id immediately. It explicitly distinguishes itself from generate_video ("not how you make a video look different (that is a new generation)"), so an agent can pick it over the generation siblings without opening a 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 WHEN clause enumerates trigger phrasings (upscale, enhance, sharpen, increase resolution) and the exclusion is explicit. It also routes the follow-up work precisely: do not call show_generation, then call wait_for_generation with the job id and channel 'media', with an explicit prohibition on get_* polling, timers and background shells.

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