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

With annotations declaring readOnlyHint=false, openWorldHint=true and idempotentHint=false, the description still adds real signal: the operation is free, it is asynchronous, and the result must be retrieved by polling list_creations, where it arrives as data. That polling instruction is exactly the kind of context annotations cannot express. It omits processing time and what happens on failure/duplicate submission, which idempotentHint=false makes relevant.

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 payoff (what the breakdown contains), then the alternative route, then the retrieval mechanics — a sensible priority order for an agent. It is dense but each clause carries information; the fragments ('Free.') are terse rather than wasteful, though the punctuation style makes it slightly harder to parse than prose.

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?

No output schema exists, so the description must convey what comes back — and it does, both by listing the analysis components and by explaining that the result is data retrieved via list_creations. The main residual gap is the mechanics of the async handoff (does the call return a job or creation id? how long does processing take?), which is only gestured at with 'Poll list_creations'.

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% and the schema description for 'video' already spells out the accepted link forms (CartViral /u/ link or direct public https file link, downloaded once). The description adds nothing further about the two parameters and never mentions 'language', so this is the baseline case where the schema does all the work.

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 enumerates concrete deliverables — hook mechanism, beat timings, spoken/shown content, on-screen text style, measured cuts — so an agent knows this is a structural analysis of a video rather than a creation tool. The verb is implicit (the title carries 'Break down'), and the opening phrase 'Why a video works' is a bit indirect, but the resource and scope are unmistakable and clearly separated from remake_from_reference.

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?

It names the sibling alternative explicitly ('To make the same structure with your product or without one, use remake_from_reference'), which is a genuine routing rule. It also flags 'Free' as a selection factor. It stops short of stating exclusions (e.g., don't use for videos you don't have a direct file link for), so it is clear context rather than a full when/when-not matrix.

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