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

No annotations are provided, so the description carries the full burden of behavioral disclosure. 'Values discarded' does disclose that values are not returned, which is useful. But the description is silent on whether nested keys are counted, what object is counted when all parameters are optional and described as 'discarded,' what errors look like, and why the json parameter is framed as validation input rather than the object to count. The mismatch between the tool's stated purpose and the schema's parameter descriptions adds ambiguity instead of transparency.

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 two short sentences (7 words) with the core operation front-loaded and zero filler. Both sentences earn their place. However, at this scale of context — 9 parameters, 27 siblings, no annotations — the terseness verges on under-specification, so it does not quite earn a 5.

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?

Given 9 optional parameters, no annotations, no output schema, and 27 siblings, the description is completely inadequate. It does not state what the return value is (beyond an implied count), what happens when called with no arguments, which parameter (if any) is the input, or how this differs from validate-json. An agent cannot determine how to invoke this tool correctly — it would not know whether to pass json, any other parameter, or nothing at all.

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 the baseline is 3, but the schema descriptions themselves are misleading: every parameter ends with 'discarded after...' (e.g., 'Git ref name; discarded after the shape check,' 'Search text; discarded after the length check'), reading like leftovers from other tools. The json parameter is described as validation input, which contradicts the description's counting purpose. The tool description adds no clarification about which parameter supplies the JSON object, so the overall semantics actively confuse rather than inform.

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 names a concrete operation — 'Count keys in a JSON object' — with a clear verb and resource, and adds a meaningful scope line ('Values discarded'). It is not a tautology and is conceptually distinct from validate-json. However, it is vague about the input source: the schema's only JSON-carrying parameter (json) is described as 'JSON text to validate; discarded after the check,' so an agent cannot tell which object is being counted or whether the json parameter even matters.

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?

No usage guidance is provided. With 27 siblings including validate-json, hn-front-count, search-query-len, and domain-shape — tools that also perform checks/counts — there is no statement of when to choose this tool over them, and no when-not or alternative conditions. The agent is left to infer relevance purely from the name and description.

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