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

Annotations only say readOnlyHint=false and destructiveHint=false, so the description must carry the behavioral load, and it adds real value beyond them: it navigates around obstacles and returns 'once you are there, or stopped', signalling that the call is blocking and may terminate before reaching the target. It does not explain what causes a stop, whether progress is partial, or any cost per move, which keeps it 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?

Three short sentences with no filler: capability and addressing modes first, then return/termination behavior, then the disambiguation to collect_bill. Each sentence earns its place and the routing hint is placed last where it does not obscure the primary purpose.

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 four-parameter movement tool with no output schema and only minimal annotations, the description covers the essentials: destination modes, obstacle avoidance, and termination semantics. It omits failure conditions, whether person and coordinates can be combined, and what the result reports, which are the remaining gaps an agent would hit in edge cases.

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 25% (only 'person' is documented, as 'A name, as look gives it'). The description compensates partially by clarifying that layer is only needed when the destination is not on your own layer, which is genuine meaning not in the schema. However it never explains the row/column grid semantics, the 0-63 bounds, or whether 'person' is an alternative to coordinates or combinable with them, leaving a real gap for a low-coverage 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?

The description gives a specific verb and resource ('walks you to a place in the room ... or up to a person') and enumerates the accepted addressing modes (row/column, layer, person). It also names a sibling it overlaps with (collect_bill) and the resource that distinguishes them, so an agent can separate the two without reading schemas. Minor ambiguity remains about whether 'place' is a grid cell or an object location.

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?

It provides a concrete when-to-use-elsewhere rule: for a dollar bill, use collect_bill, which walks and collects. That is an explicit alternative with a selecting condition. It does not address other plausible alternatives in the sibling set (travel, use_door, leave) or state when walking is inappropriate, so it stops short of full routing guidance.

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