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

Annotations already declare readOnly/openWorld/destructive=false, so safety is covered; the description adds genuinely non-obvious behavior — that a call with no `category` returns no records at all but a catalog in `windows`. It also discloses the `top` cap of 10 and the unit/format conventions for pace. It does not explain anything about date scoping or the `athlete` parameter, which is the remaining behavioral gap.

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?

Front-loaded with purpose, then the two-step workflow, then per-parameter notes — a sensible order for an agent. Density is high and each parameter line earns its place. The pace-formatting aside (238 s/km is 3:58 per km) is useful but is presentation advice sitting in a tool definition, slightly padding the tail.

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?

An output schema exists, so return-shape detail is rightly omitted, and the one non-obvious return case (empty call yields the catalog) is covered. For an 8-parameter analytics tool with a zero-coverage schema, the undocumented `athlete` and date-range parameters are the notable omissions — particularly `athlete`, given list_athletes is a sibling and multi-athlete scoping is ambiguous.

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?

With 0% schema coverage the description carries the full burden, and it mostly delivers: it enumerates the six `category` codes with their units, gives both legal `axis` values with meaning, states that `windows` are seconds or metres and must come from the catalog, and caps `top` at 10. Three of eight parameters — `athlete`, `date_from`, `date_to` — receive no mention at all, so it stops short of full compensation.

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-plus-resource ('Read personal records') and immediately defines the resource precisely: best power, pace, heart rate, torque, vertical speed or cadence held over a duration or a distance. That is distinguishable at a glance from every sibling (get_stats_chart, get_training_load, list_metrics), none of which return all-time bests.

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 prescribes an explicit call sequence: invoke without `category` first to get the category/window catalog, then read records with a `category`. It also names two prerequisites by sibling tool ('windows ... taken from the catalog only', 'sports: spt_… handles from list_sports'), so the agent knows where dependency values come from.

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