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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, yet the description still adds real behavioral context: agreement is defined as vote-match with the settled outcome and explicitly 'not accuracy', pending data is never read, percentiles are computed only among non-system graders at or above min_n, and seal timing is network-wide and 'never per grader'. These are non-obvious traits an agent could not infer from the annotations.

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 front-load is good ('Advisory, unranked grader patterns from settled jury seats in a segment'), but the remainder is a single run-on enumeration of nine metrics with parenthetical definitions, making it hard to scan. The information is mostly load-bearing, but the packaging is dense and non-paragraph-structured.

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?

With no output schema, the description carries the burden of describing the return shape and does so field by field, plus medians/bands and the CSV endpoint. Some returned terms remain unexplained ('closes', 'score mix', 'distribution bands'), which leaves minor gaps for a complex 7-parameter analytics tool.

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 schema already documents the seven parameters. The description still adds meaning by explaining that min_n gates the percentile population, not just row inclusion, and that segment 1 differs behaviorally (agreement and peers only). Marginal but genuine added value over the schema.

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 a specific resource (grader patterns from settled jury seats in a segment) and a specific scope (advisory, unranked). It implicitly separates itself from ranking-oriented siblings like get_governance_graders or get_leaderboard by stating it is 'not to rank them', though it never names an alternative tool. The dense enumeration of metrics muddies the one-line takeaway.

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?

'Use it to see how graders' settled votes and scores spread, not to rank them' is an explicit when-to-use plus a when-not. The segment caveat ('Segment 1 has no settled scores: agreement and peers only') and 'Pending answers and scores are never read' further bound expected usage. It stops short of naming which sibling to use for ranking instead.

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