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

Annotations only declare readOnlyHint=true, so safety is already covered; the description's real contribution is the billing model, '$0.005 USDC; not charged if nothing matches', which the agent cannot get from structured fields and which materially affects whether it should call this speculatively. It does not disclose result shape (item fields returned), whether results are capped by limit, or any rate/latency behavior, so the added context is valuable but partial.

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?

Two tight sentences: capability plus example first, then constraints and pricing. Every clause carries information, though the sort enumeration slightly restates the schema enum and the macro filter list echoes the property names, so a small amount of the text is redundant rather than additive.

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-parameter, single-required-parameter query tool with no output schema, the description covers intent, filters, sorting, and cost but omits what a result actually looks like and how the default limit of 5 shapes the response. Nothing is misleading, but an agent cannot predict whether it receives 5 items or all matches, which matters for follow-up calls.

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 57%, and the description compensates well: it names all four macro filters (max calories, min protein, max carbs, max fat) and expands the sort_by enum into plain language ('protein, calories or protein per calorie'), clarifying that the last is a ratio rather than a raw value. The 'limit' parameter – which is bounded 1-10 with a default of 5 – is never mentioned, leaving one meaningful knob to the bare 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 pairs a specific verb and resource ('Find menu items at a restaurant') with the defining scope 'that fit macro targets', then anchors it with a concrete example (highest-protein items under 600 calories at Chick-fil-A). It stops short of naming how it differs from get_restaurant_menu, so an agent must infer the boundary, but the filtered/sorted nature of the query is unmistakable.

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 worked example supplies clear usage context for the canonical case: a user with macro constraints wants ranked candidates, not a raw menu. There is no explicit exclusion ('don't use this when you just want the full menu – use get_restaurant_menu') and no mention of the sibling tools, so routing is left to inference.

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