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

Annotations already establish that this is a read-only, idempotent, closed-world computation, so the description correctly spends its words elsewhere. It adds real behavioral detail: that results include per-inspection attribution ('which inspections drive each measure') and time-based decay behavior ('when they step down and drop off'), and it forecloses a common false expectation about percentiles. It does not describe error handling for malformed inspection records or defaults beyond asOf, 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?

Two sentences, front-loaded with the core capability, with the negative scope ('Percentiles are not computed') placed deliberately at the end as a qualifier. The parenthetical input format is dense but earns its space by telling the agent what shape of inspection data to supply. Some of that parenthetical detail duplicates the schema.

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?

There is no output schema, so the description carries the burden of describing returns and does so at a useful level (measures per BASIC, driving inspections, step-down/drop-off timing) while explicitly ruling out percentiles. For a computation tool with three parameters, this is close to sufficient; a brief note on required field completeness in each inspection object would have closed the remaining gap.

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 description adds meaning the schema does not: the severity scale is 1-10, the enforcement-relevant role of outOfService and level, and the semantic linkage of each violation's basic field to one of the four output measures. The carrierType and asOf parameters are still only explained by the schema, which is why this is not a 5.

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 opens with a specific verb+resource ('Compute FMCSA SMS measures') and enumerates the four exact BASIC categories (HOS Compliance, Vehicle Maintenance, Driver Fitness, HM Compliance) plus the input it derives them from (roadside inspections). It also states a scope boundary ('Percentiles are not computed'), so an agent can distinguish it from percentile/SMS-lookup style tools in the sibling list without opening the 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?

Usage context is clear: this is the tool for turning inspection records into BASIC measure scores and for seeing which inspections drive each measure. The explicit exclusion of percentiles acts as a when-not statement, steering agents away from expecting a full SMS snapshot. It stops short of naming an alternative tool for percentile computation, so it falls one step below full 5-level guidance.

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