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

Adds substantial context beyond the annotations: the consent restriction (no public figures/celebrities), that the call spends tokens, the priced-ask/confirmation flow, card/URL handling, and the idempotency-key replay semantics. These are genuine behavioral disclosures a schema cannot carry. It loses a point for not stating what the job produces or how long it typically takes, but with destructiveHint=true and openWorldHint=true already declared, the bar is met.

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 and telegraphic, with labeled sections (CONSENT, THE CARD, WAIT, SPENDS TOKENS, IDEMPOTENCY) that make scanning easy. It is dense to the point of terseness in places (e.g., 'Ask in the SAME message as the price'), but nearly every clause carries actionable instruction.

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 costly, destructive, open-world generation tool with no output schema, the description covers the critical lifecycle: recipe lookup, pricing, consent, card suppression, waiting, and idempotency. What remains thin is the semantics of the two modes and what comes back from the call, which the agent will only learn from other tools.

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?

At 67% schema coverage the description usefully compensates: it sets the `resolution` default ('use `720p` unless the user asked for `1080p`'), restricts `model` to explicit user request, and explains `estimate_only` and `idempotency_key` behavior. The main gap is `mode` (swap vs motion) and `reference_images`, whose meaning as 'the user's subject' is only gestured at rather than defined.

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 gives a specific-ish verb and resource ('Remakes a clip with the user's subject'), but it never explains the two `mode` values (swap vs motion), which is the core distinction of the tool. It also does not say how it differs from close siblings like `motion_control`, `edit_video`, or `generate_video`. A reader can tell roughly what it produces but not which of the two operations they are choosing.

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?

Strong operational routing: 'Read `get_recipe` `reshot` first', do not call `show_generation`, use `wait_for_generation` instead of `get_*` polling, price with `estimate_only: true` before spending. It gives clear when-to-use and when-not behavior. It stops short of the 5 because it never contrasts this tool with its generation siblings (motion_control, extend_video, edit_video).

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