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

Annotations already declare readOnly/idempotent/non-destructive, and the description adds value beyond them: 'Costs NO credits', it performs an authenticated balance check (an auth-adjacent side effect), the `recent` block covers the last 30 days of real runs with fail rate, auto-judge score, ratings and median render time, and candidates are explicitly not selectable. These are behavioral facts not derivable from the annotations.

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 correctly and zero fluff sentences, but it is a single very dense sentence packing modes, limits, stats, cost, auth and defaults together, making it hard to scan. Some of the selection guidance could be separated from the return-value inventory.

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?

There is no output schema, so the description carries the burden of describing return contents, and it does so thoroughly (per-mode inputs/limits/quality/credits, pick-when, latency, 30-day run stats). Nothing an agent needs to decide whether and how to call it is missing.

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?

The only parameter, include_candidates, is already 100% covered by the schema and the description merely restates it. The instructions about passing the id as `model` or `model:"auto"` actually describe parameters of the generate_* tools, not of list_models, which risks confusing an agent about this tool's own signature.

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+resource ('List the generation models agent-media can use') and enumerates precisely what is returned per model: modes, exact inputs, limits, aspects, qualities, credits/second, pick-when line, latency, selection guidance, plus 30-day run statistics. This is clearly distinguishable from siblings like quote, get_account, or the generate_* tools.

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?

Explicitly states timing and alternatives: 'Read this BEFORE choosing a model or a mode for generate_video / generate_image / generate_audio', plus compare the included balance with quote before generation, and use include_candidates:true only when planned models are relevant. It also names the default (seedance-2.5) so an agent knows when it can skip this call.

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