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

Annotations only say this is a non-readonly, non-idempotent, non-destructive write. The description adds material context the annotations do not: a prerequisite skill load, constraints on fps/splitting, and a precise cost model (~5.7 credits per minute of range, billed by the second). It does not describe what happens after the call (job lifecycle, return value), keeping this from a 5.

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?

Four dense sentences, front-loaded with the core action, then constraints, cost, and prerequisite. Almost every clause carries load; it is slightly packed but no sentence is filler.

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 7-required-param, no-output-schema mutation the description supplies the crucial operational context: prerequisite skill, cost, and fps/clip constraints. It omits post-call behavior (whether it starts a job needing wait_for_jobs) and the meaning of assetId, leaving a small but real gap.

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 only 14%, so the description must compensate, and it does for the non-obvious parameters: 'detection' is defined as a numbered detection from the latest find_objects result, fps is pinned to normally 8, and label and the source-time range (startTime/endTime) are named. assetId is left implicit and projectId relies on the schema, but the key ambiguous fields are covered.

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 (track) and resource (a numbered detection) and pins the source ('from the latest find_objects result on this asset'), which immediately distinguishes it from track_faces, track_box, and track_points. An agent can identify the correct tool without opening any schema.

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 concrete operating guidance: specify the source-time range, fps (normally 8), and label, don't lower fps, don't split a long clip, and load builtin-object-tracking first. It routes implicitly from find_objects but never explicitly contrasts with the sibling track_* tools, so it stops short of a full when-to-use statement.

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