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

Annotations only declare the safety profile (openWorld, non-destructive, not read-only), while the description adds genuinely operational detail: the call blocks for about two minutes, may return an unfinished generation with status queued/running, and the successful result carries a CDN URL in Assets. That is exactly the behavioral context the annotations cannot supply.

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?

Three tight sentences in a logical order: what it makes, where the artifact lands, then the job/blocking semantics and fallback. No filler or redundant restatement of the name.

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?

With no output schema, the description correctly compensates by describing the return shape (generation record with status, CDN URL in Assets) and the timeout/polling contract. Combined with a fully documented input schema, an agent has everything needed to call it and handle the async case.

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 100% and the parameter docs are unusually rich (tier vs model/provider cancellation warnings, vendor-declared params, unknown keys ignored), so the schema carries the load. The description says nothing about tier, model, params, or provider, adding no meaning beyond the schema — baseline 3 applies.

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 ('Generate an image') and narrows the domain with examples ('icon, sprite, art') for a title, which distinguishes it from the adjacent generate_video/generate_music/generate_audio/generate_text/generate_three_d siblings. It also names get_generation as the polling counterpart, so the agent can place the tool precisely.

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?

Gives clear usage context: the output is meant to be referenced from title config or code via the CDN URL in Assets, and it states the alternative path when the job isn't finished (poll get_generation with the id). It does not state when to prefer a different generation tool instead, but the surrounding guidance is strong.

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