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

Annotations already declare readOnly/idempotent/non-destructive, and the description adds real behavioral context beyond them: the interim default (an image takes the best tier), the null selected_tier for clips awaiting a decision, and the fallback when a tier is unaffordable. It does not cover latency or whether benchmark data is cached, so it stops just short of full transparency.

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?

Every clause carries decision-relevant information — pricing, defaults, the ask-once protocol, the affordability fallback, and the recipe preference — and the core purpose is front-loaded. The cost is that it arrives as one dense run-on paragraph, which makes the decision rules harder to scan than they need to be.

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?

An output schema exists, yet the description still names the key fields an agent must branch on (affordable, free_now, selected_tier, image_priority/clip_priority), and it covers the full decision workflow including the ask-user-once policy and recipe handoff. Nothing an agent needs to call this correctly 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?

Schema coverage is 0%, so the description must carry the load: it fully explains `task` ('also get the catalog recipes made for that job', with the recipe follow-up path) and ties `kind` to priorities and selected_tier. It only illustrates kind with image and clip, leaving video/tts/music behavior to inference from the enum.

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 with full scope: 'Which model to use for a job, in three tiers (best, middle, cheapest), priced for this account', and enumerates the returned signals (credits per reference request, affordable, free_now, priorities, selected_tier). It is clearly distinguishable from siblings like list_models or quote_generation because it is account-priced and returns a selected tier per task.

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 instructs 'Call it before choosing a model and take the selected tier', defines what to do when no priority is saved (ask once at first request of that kind, then persist via set_preferences, never ask again), what to do when a tier is unaffordable (take the best affordable and say so), and what to do with recipe results (prefer generate_with_recipe / start_recipe_run). Both alternatives and conditions are named.

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