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

Annotations already declare readOnly/idempotent/non-destructive, but the description goes further by documenting the non-throwing error contract ({"error": ...}, with a concrete example), the typical input/output payload, and the retry-safety guarantee after correcting input. It also flags the silent-failure hazard (boxes that "quietly ruin every metric"), which helps the agent understand the cost of bad input.

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 purpose, then format semantics, then a concrete example, then routing rules and error behaviour — a defensible order. It is longer than strictly necessary: the paragraph on how the three formats disagree overlaps with the enum descriptions already in the schema, and the example return value duplicates what the output schema covers.

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 read-only, idempotent conversion tool with a full input schema and an output schema, the description supplies everything an agent needs: when to call it, what the payloads look like, how failures are surfaced, and that retries are safe. No gap is left that the structured fields don't cover.

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%, so parameter documentation is already strong. The description adds a worked example that shows all five meaningful fields in use together (including the image_width/image_height requirement when YOLO is on either side), which is more than restating the schema. It stops short of explaining the `clip` flag's behaviour, which is left entirely to the schema.

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 and resource ("Convert bounding boxes") and enumerates the exact three formats involved, so an agent knows the operation's scope immediately. It also names the sibling tools it is not (detection_metrics, nms), making it distinguishable 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?

States the triggering condition ("whenever a dataset and a model disagree about format") and explicitly excludes two sibling tools by name with their purposes. Nothing is left for the agent to infer about when to route here.

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