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

Annotations already declare readOnlyHint, idempotentHint, and non-destructive behavior, so the description doesn't need to repeat those. It adds meaningful behavioral detail beyond annotations: the meaning of 'DUE' as arriving now, the inclusion of a delay flag and vehicle id, and the hidden behavior that find_stop performs a stop lookup for the caller.

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 three well-ordered sentences plus a concrete example. It front-loads the core function, then covers the two parameter modes, and ends with a call example. No sentence is wasted and the structure makes the content scannable.

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 prediction tool with no output schema and six optional parameters, the description covers the essential aspects: what the tool returns, the DUE convention, how to identify a stop, and how to invoke with a stop name. The max and _apiKey parameters are documented in the schema, and the description need not restate them. Slight room remains for explicitly noting comma-separated stop_ids, but that is covered in the schema and not critical for invocation.

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 input schema already has 100% description coverage, so the baseline is 3. The description adds practical semantics on top: stop_id is the '4-5 digit stop number posted on CTA bus-stop signs', find_stop is a 'street/intersection fragment', and the route parameter is required when using find_stop. The included example call further clarifies how these parameters combine.

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 states a specific verb and resource: 'arrival predictions at a bus stop' for the Chicago CTA bus tracker, and enumerates the output fields (route, destination, predicted minutes, DUE, delay flag, vehicle id). This clearly separates it from sibling tools like cta_bus_positions and cta_train_arrivals, since it is explicitly bus predictions at a stop rather than positions or train arrivals.

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 gives clear usage context by explaining two invocation modes: pass stop_id optionally with route, or pass route + find_stop when only a stop name is known. It does not explicitly list when to prefer sibling tools, but the stop-based prediction scenario is well defined, and exclusions are unnecessary given the read-only nature.

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