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

Annotations already indicate non-read-only and non-idempotent nature. The description adds crucial behavioral context: the run is asynchronous (you must poll get_run until is_terminal), credits are charged only on completion, and device selection constrains the output. It discloses the charging condition and the need for polling, which are not in annotations, making the side effects transparent.

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 description is compact: two sentences pack the purpose, the return value, the polling requirement, and the charging rule. It is front-loaded with the primary action, and every clause contributes new information. No filler or redundancy.

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 tool with 7 parameters, an output schema, and complex side effects, the description covers all essential operational aspects: how to start, what to do with the returned id, device multiplicity, and when billing occurs. The existence of an output schema relieves the description from detailing return structure, and the description is complete enough for correct usage.

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?

The schema covers 100% of parameters with meaningful descriptions, so the baseline is 3. The description adds value by explaining relationships: the default guide is the newest, device variants (desktop/phone/tablet) and the rule that one run equals one device, plus how to handle multiple devices. This goes slightly beyond the schema's per-parameter explanations.

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 clearly states a specific verb ('Start recording') and a specific resource ('a video for a project'), and further specifies the object through guide_id and device. It differentiates from siblings by describing its role in the run lifecycle (creating a run, returning an id to poll) and mentions the asynchronous nature. This leaves no ambiguity about what the tool does.

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?

The description provides direct usage guidance: it explains that one run per device must be started separately for desktop and phone versions, and instructs to poll get_run until is_terminal. It implies the context of starting a recording but does not explicitly state exclusions or alternative tools. The instruction on when to use (when you want to record) is implicit but clear enough.

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