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

Annotations already establish readOnly/non-destructive, and the description goes well beyond them with substantive behavior: missing windows are genuine gaps and are never interpolated, series points carry each window's own values rather than being aggregated, percentiles are withheld as null unless n×(1−q) ≥ 3 (with concrete thresholds p50 ≥ 6, p95 ≥ 60, p99 ≥ 300), and 'flagged' means p50 ≥ 10× the median of per-rule p50s within the group. It even states a deliberate non-feature — absolute thresholds are intentionally unsupported — which prevents a whole class of misuse.

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 the identity of the tool and its output shape, then progressively narrower caveats. Each clause carries real information — no filler or restated name. It is very dense in the back half (percentile gating thresholds and the flagged rule packed into single clauses), which slightly hurts scanability, but nothing is wasted.

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 must explain the return payload — and it does, covering both merged distributions and the per-window series, null semantics, gap handling, and the flagged marker. With all parameters documented and safety covered by annotations, an agent has everything needed to call this correctly and interpret the result. Only minor omissions (range limits, series point limits) remain.

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 every parameter (from/to defaults, ruleId provenance, groupId, versionId defaulting to live) is already documented in the schema itself. The description contributes no additional parameter-level meaning — it does not mention time-range defaults, the live-version default, or any constraint on window length versus range. Baseline 3 applies when the schema carries the parameter burden.

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 opening clause names the resource precisely — single-rule latency detail — and enumerates the two payload shapes returned: merged phase × cacheState distributions and a 60s-window time series. The word 'single-rule' implicitly separates it from the aggregate/overview sibling, and 'ruleId (from lexq_profile_overview)' reinforces that relationship. It lacks an explicit verb and never names the alternative tool to use when a whole group is wanted, which keeps it from a 5.

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 is only implied: by calling itself 'single-rule' it hints you should not use it for group-level or multi-rule analysis, and the parameter hint points to lexq_profile_overview as the way to discover ruleId. There is no explicit when-to-use / when-not-to-use statement, no guidance on choosing time ranges, and no mention of what tool to reach for instead for aggregate views.

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