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

Goes well beyond the readOnly/openWorld annotations by disclosing the data source (`source: "captions"`), the exact failure mode and status code for uncaptioned videos, and the billing rule (1 credit, only charged when a transcript is actually returned). Missing only details like whether cached results still cost a credit or how long transcripts can be.

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 scope and ID/URL constraint are front-loaded and no sentence is filler; the credit rule and the 501 caveat each earn their place. The opening fragment 'Ad video transcript.' duplicates the title and the ID/URL exclusivity is stated once at the end rather than with the other parameter details, but overall it is tight.

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 a 4-parameter tool with no output schema, the description covers the selection constraint, cost and failure path, which is the important part. But with `cache_max_age` undocumented anywhere and no hint about the shape of a returned transcript (plain text vs. timestamped segments), an agent still has to guess at both caching behavior and what it will receive.

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 75%, so the baseline is a 3, and the description does add one genuinely new constraint not expressible in the schema: 'Pass exactly one of `id` or `url`' (mutual exclusivity). However, it adds nothing about `language` or the entirely undocumented `cache_max_age` parameter, so the coverage gap is only partly compensated.

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: 'Retrieves a transcript for a single Facebook Ad Library video ad by ID or URL', and narrows the mechanism to Facebook's published caption track. This clearly separates it from facebook_ad (ad metadata) and from post/reel transcript siblings on other platforms, though it never names those alternatives explicitly.

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 a concrete when-not condition: ad videos without captions return `not_configured` (501) and are not charged, so the agent knows this tool silently fails for that class of input. It does not route the agent to any sibling (e.g. facebook_ad to first confirm the ad is a video) or state prerequisites, so it stops short of full when/when-not/alternatives coverage.

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