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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, closed-world), so the bar is lower, yet the description adds genuine traits: no target execution or assessment occurs, pagination must restart if the retained population changes, and a metadata projection preserves real nextCursor/total. These are non-obvious behaviors the annotations do not convey, though the final sentence about 'exact authorized API resource' is opaque.

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 purpose is front-loaded and the text is short, but it is dense and jargon-heavy: 'A metadata projection preserves the actual nextCursor/total; omitted items are available from its exact authorized API resource' is convoluted and hard to act on. Some sentences earn their place, but the projection sentence does not read cleanly.

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 does usefully describe pagination stability and the projection of nextCursor/total, and it clarifies this is a pure read with no execution. But the total absence of parameter documentation for a 6-param tool leaves a clear gap for correct invocation.

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 description coverage is 0% across 6 parameters, so the description must carry the burden and it largely does not. It references baseline/candidate broadly and mentions cursor/total via the projection, but never explains limit, cursor format, or how the four required IDs (projectId, environmentId, baselineBatchId, candidateBatchId) relate. It only partially compensates for the coverage gap.

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 sentence gives a specific verb ('Page') and resource ('every comparison case') scoped to baseline/candidate run references, including the edge case of missing/incompatible pairs. It separates this from the evidence-reading siblings by pointing to 'report reads' for either run's evidence. A little more crispness about what a 'comparison case' is would push it to 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?

It explicitly redirects evidence lookups to report reads ('Use report reads for either run's evidence'), which is a real alternative-tool routing. It also gives a usage constraint for pagination ('Restart at the first page if the retained population changed'). However, there is no explicit when-not guidance and no statement of prerequisites for the batch IDs.

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