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

Annotations already declare readOnlyHint, idempotentHint and openWorldHint=false, so safety of repetition is partly covered; the description nonetheless adds real value with the ~45 s bound and "never charges", a billing guarantee nothing in the structured fields conveys. It does not say what a timed-out or failed job looks like on return, which would be the remaining behavioral gap.

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?

Three clauses in roughly two sentences: what it does, how safe it is to repeat, and what comes back — front-loaded with the primary purpose and with no filler. Every phrase carries information an agent would otherwise have to guess.

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?

With an output schema present, the description need not enumerate return fields, and it still summarizes them usefully (download URLs, preview, size in metres, triangle count, asset id) while covering timeout and cost behavior. What remains missing is the non-success path — what is returned or raised when the wait expires or the job errors.

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 0%, so the description must carry parameter meaning, and it only partly does: "up to about 45 s" gestures at wait_seconds and its default, but never says it is caller-configurable (0–50, minimum 0) or that wait_seconds=0 means a pure non-blocking status read. job_id is left entirely implicit.

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?

"Waits up to about 45 s for a job to finish, or reads where it stands" gives a specific verb (wait/read) and resource (job) plus the two modes of operation, so an agent immediately understands this is a status/polling tool. It does not explicitly name a sibling (generate_model, get_asset) to contrast with, which keeps it short of a 5.

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?

"Safe to call again and again" and "usable as input to the next step" spell out the intended polling loop and downstream handoff, which is the core usage context for a job tool. It stops short of naming what to do instead when the job never finishes (or of pointing to get_asset once an id exists), so no explicit exclusions.

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