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

Annotations already declare readOnlyHint=true and destructiveHint=false, so the safety profile is covered. The description adds meaningful behavioral context: every optional constraint must be sent as null (never a placeholder), and the tool must not be treated as a quote, generation command, or acceptance guarantee. This clarifies limitations beyond what annotations provide, though it does not describe the return format or whether the tool can ask follow-up questions beyond the hint to ask about missing goals.

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?

The description is five sentences, each earning its place: the trigger condition, the null rule, the prerequisite to call calculate_project_budget, and two exclusion clauses that disambiguate from sibling tools. It is front-loaded with the core purpose and then adds operational constraints, with no filler or repetition of schema content.

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 13-parameter tool with no output schema and a large sibling set, the description covers the key decisions: when to use, what to ask, how to handle nulls, and which sibling to delegate cost calculations to. It does not explicitly describe the output structure (e.g., a ranked list with alternatives), only implying it through 'best-fit ... plus strong alternatives', which is a minor gap given the absence of an output schema.

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?

With 100% schema description coverage, the baseline is 3, but the description adds cross-cutting parameter semantics: the universal null-for-unstated rule applies to all 13 optional parameters and prevents placeholder misuse (e.g., 'never use 300 as a placeholder' in maxDurationSec). It also reinforces the relationship between budgetCeilingCents and calculate_project_budget, and ties priorities like lower_cost to that sibling tool, providing guidance that goes beyond per-field schema text.

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 states a specific verb ('recommend') and resource ('MaxVideoAI models') and defines the exact trigger: the user is undecided, asks for advice, or wants a best-fit model plus alternatives. It further distinguishes itself from siblings by explicitly excluding validation, pricing, and execution use cases, and by naming calculate_project_budget as the alternative for cost questions.

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?

The description provides explicit when-to-use ('user is undecided, asks for advice') and when-not-to-use conditions ('already chose a compatible model and only wants validation, pricing, or execution'; 'not an exact quote, a generation command, or a claim that a provider will accept a job'). It also names the sibling tool calculate_project_budget and instructs to use it before requesting a cheaper/lower-cost alternative, leaving no ambiguity about routing.

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