get_verdict_divergence

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

With no annotations provided, the description fully carries the behavioral disclosure burden. It discloses that only confident rows surface (uncertain middle band is dropped), rows are sorted by confidence so strongest calls come first, each section is paginated with filtered_total, and that group_sufficiency divergences are included. This gives the agent a clear model of the tool's behavior and limitations.

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 description is organized with clear labels (missing_mapping, spurious_mapping, group_sufficiency), uses bolded terms for key concepts, and front-loads the core purpose. Despite its length, every section adds necessary operational detail—actionability, confidence sorting, pagination, and follow-up tools—so there is no wasted text.

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?

The description is complete for this complex tool: it covers all divergence kinds, explains their directionality, describes pagination and filtered_total for full review, mentions the related accept/dismiss tools, and notes that group_sufficiency rows are observation-only. The output schema exists and the input schema is well-described, so no additional return-value documentation is needed.

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?

The schema already documents most parameters well, including kind values, limit clamping (1-500), and include_dismissed semantics, so the baseline is 3. The description adds meaningful context beyond the schema by explaining what 'confidence' means via p_covers, how missing_mapping/spurious_mapping translate to add/remove actions, and why pagination matters. This elevates it above the baseline, though server_version and model_id are not elaborated in the description.

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 statement of what the tool does: it shows where LLM verdicts disagree with the model's authored state. It then clearly distinguishes the two coverage divergence kinds (missing_mapping and spurious_mapping) and mentions group_sufficiency divergences, making its scope unambiguous and distinct from sibling tools like accept_coverage_divergences and dismiss_verdict_divergences.

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?

The description provides explicit guidance on when to use the tool and how to act on its results: confident missing_mapping rows should be accepted to ADD mappings, confident spurious_mapping rows should be accepted to REMOVE mappings, and group_sufficiency rows are observation-only. It also names the related tools (accept_coverage_divergences, dismiss_verdict_divergences) and explains pagination strategy for handling large result sets.

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