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

With annotations present but sparse, the description adds substantial behavior: this is an async queued job with a job-view polling pattern and an output_id return, Highlights uses default-report goals/quotes and original audio with no artificial narration, and the prompt cannot select different quotes. One wrinkle: idempotentHint is false while the text says to retry with the same key, which is meaningful but explained only implicitly (dedupe via idempotency_key rather than natural idempotency).

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 purpose and the reuse rule are front-loaded, and the polling/retry instructions are compact. Later clauses are telegraphic (e.g. 'with original audio when available and no artificial narration for chat studies') and require a re-read, but no sentence is filler.

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 an async generator with an output schema, the description covers the queuing model, how to poll, where to read the result, idempotent retry, and output-type-specific behavior. The remaining gap is semantics for the optional parameters (template_id, name, include_synthetic), which an agent would otherwise have to guess at.

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 13%, so the description must compensate and does so only partially: it conveys the three type values, the limits of prompt for Highlights, the retry role of idempotency_key, and implicitly session_id. name, template_id, include_synthetic, and organization_id are left entirely to the schema, so several parameters remain semantically thin.

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 first sentence gives a specific verb ('Queue') and three concrete output resources (report, native PowerPoint, saved Highlights video) scoped to a Hub session. It is clear what the tool produces, though it does not explicitly distinguish itself from close siblings like generate_report or queue_report_slides.

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 supplies a real usage rule: it reuses saved reports and findings so exhaustive analysis need not run first. It also gives the operational loop (poll inspect_research_session view=job, then read output_id) and a retry rule. It stops short of naming which sibling to use for the non-reuse or slide-specific cases.

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