get a generation

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

Annotations already cover the safety profile (readOnlyHint true, destructiveHint false), but the description adds substantial behavior the annotations cannot: the waiting contract of roughly 50 seconds, the instruction to re-call with the same id rather than resubmit, and the billing consequence that a new create starts a separate billed generation. It also warns that outputMediaUrl is signed and must be copied verbatim. The only minor gap is that it does not state whether reads are rate limited under repeated polling.

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 with the purpose and the returned fields, then segmented into labeled blocks (usage, Not for, Waiting, Result) that each carry distinct information. It is long, but the length is justified by a polling tool with resubmission and signed-URL pitfalls; the deprecated outputUrl alias note is the one line that could be trimmed.

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?

Although an output schema exists and return values need not be re-explained, the description adds the status transitions, the verbatim-copy requirement for the signed URL, and the failure-path instruction to report error and errorCode. For a read/polling tool with four parameters, nothing an agent needs in order to call it or handle its result 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 100%, so the baseline is 3, and the description earns above baseline by specifying the interaction rule between two parameters: pass wait: true and leave timeout out, and call again rather than sleeping. That combination rule is not spelled out as such in the schema, and the warning about never resubmitting adds meaning to the id parameter.

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 ('Read one sync. generation by id') and immediately enumerates what is returned: status lifecycle values, outputMediaUrl, error and errorCode. It explicitly differentiates itself from jobs_get/jobs_wait, which read the same generation in a generic job shape, so an agent can choose correctly without inspecting schemas.

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 the positive trigger (follow a generation by the id a create call returned, naming lipsync_create, dubbing_create, lipsync_create-timeline, dialogue-videos_create), a 'Not for' exclusion block routing to generations_list, projects_list-generations, generations_get-costs and generations_get-continuation, and an explicit alternative (jobs_get/jobs_wait). This is the when/when-not/alternatives triad in full.

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