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

Annotations already declare openWorldHint=true and non-idempotent, but the description adds contract-level detail the annotations cannot convey: the job is asynchronous, the server performs the download, a job handle is returned rather than the artifact, and raw bytes are unsupported. These are the exact behavioral traits an agent needs to plan a multi-call workflow.

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-loads what the tool does and its return contract before the parameter constraints and retrieval path. Every sentence carries new information — no filler, no restatement of the title.

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?

There is no output schema, and the description compensates by explaining what comes back (a job handle) and exactly how to turn it into the final cutout PNG, including both retrieval tools and the argument to pass. An agent can complete the whole async flow without guessing.

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 100% and both parameters are fully documented there, including the https-only and private-address restrictions and the mutual-exclusivity rule. The description largely restates the either/or constraint, adding only minimal framing, so the baseline 3 applies.

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?

Opens with a specific verb+resource ('Start a background-removal job') and immediately clarifies the async contract ('Returns a job handle immediately'). This clearly distinguishes it from synchronous siblings like audio_redaction and from the retrieval tools (get_job, get_file_download_url) it references.

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 explicit routing: use either imageUrl or imageFileUrl (exactly one), and names the follow-up tools and their arguments for retrieving the result. It also states a negative condition — 'cannot accept raw file bytes' — and names the two recovery paths (upload via web app, or host publicly).

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