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

Annotations supply only readOnlyHint=false; the description adds real behavioral context beyond that: totals are derived from the foods rather than supplied, there are per-client and per-meal caps, and the tool is confirm-gated so a default call previews rather than writes. It stops short of saying what happens when the caps are exceeded (error vs. silent truncation) or what the preview payload contains.

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 tight sentences: purpose first, then the derived-totals rule, then the limits and the confirm gate. The parenthetical "(confirm-gated: previews until confirm: true)" is slightly clunky but front-loads a critical control the agent must not miss, so it earns its place.

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 mutating tool with nested objects, no output schema, and only 50% schema coverage, the description covers the essentials: what is created, that totals are computed, the limits, and the two-step confirm workflow. Missing are the error behavior at the meal/food caps and any note on what a preview returns, which an agent would still have to discover.

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 50%, so the description has to carry weight, and it does: it clarifies that items are individual foods whose per-item calories/macros are summed into totals, and it explains the confirm flag's preview-vs-execute semantics beyond the schema's own wording. The clientId and name parameters are only implied rather than explicitly mapped, which is the remaining gap.

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 ("Save a meal ... to a client's My meals") and even defines the domain object as "a named group of foods, ingredient-itemised," which the sibling names (list_client_saved_meals, update_client_saved_meal, delete_client_saved_meal) do not. The creation verb alone distinguishes it from the update/list/delete siblings, so an agent can select it without opening another 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?

It states the motivating use case ("so they can re-log it in one tap") and the operating limits ("Max 100 meals per client, 40 foods each"), plus a clear procedural condition for the confirm gate. It does not name an alternative tool or say when not to use it, which keeps it short of a 5.

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