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

Annotations already flag destructive/non-readonly/idempotent, but the description goes well beyond them: it discloses coin deduction, that the call queues rather than returns media, the required signed quote/pricing revision/approval handle, and the idempotency-key reuse rule on uncertain retry. That is real operational context an agent cannot infer from the schema.

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 'PAID ACTION' and the consent gate, and nearly every sentence carries a distinct rule (consent, do-not-fabricate, idempotency, status-first). Dense but not padded; the single long paragraph could be broken up for scanability.

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 destructive, high-parameter, no-output-schema mutation, the description covers cost, async return shape, prerequisites and retry safety, which is the core of what an agent needs. It does not address failure/refund behavior or how to retrieve the result afterward, which keeps it short of full completeness.

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?

Description meaningfully explains the high-stakes required params (quote_confirmation, pricing_revision, expected_price, approval_handle, user_consent_text, idempotency_key), which compensates for the 26% schema coverage on those. But the majority of the 23 params (image_urls, source_urls, values, element_payloads, result_origins, durations, buckets) get no explanation in either description or schema, leaving substantial ambiguity.

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?

States a specific verb and resource ('Deducts coins and queues one durable image/video generation') plus the key return shape ('request_id rather than a finished image'). It is clearly distinct from batch and canvas siblings, though no sibling is named explicitly.

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?

Explicit gating: 'use only after explicit human confirmation of one AIMixer quote for precisely the same model, inputs and coin amount', plus enumerated acceptable consent sources and the counter-case 'Never invent the reply or treat an assistant acknowledgment as consent'. It also routes the agent to inspect status first when a prior submission may have succeeded ('first inspect status').

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