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

Annotations only declare read-only/idempotent/open-world, but the description adds real behavioral rules: omitted genes are never treated as normal/negative/No Result, every prerequisite must be explicit, source text/populations/classifications are preserved, and recommendation vs incomplete-diagnostic paging advance independently. It also warns to consult source guidelines before clinical use. Minor deduction because some added clauses ('without joins', 'retain ordered alerttext, No CDS and n/a') are cryptic rather than clarifying.

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 correctly: drug mode first, then the matching modes, then the view toggle, then invariants, paging, and the clinical caveat. Dense but nearly every sentence carries operational content for a 14-parameter, three-mode tool. A few compressed phrases cost clarity without saving much space.

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 an output schema present, return values need not be explained, yet the description still points to page.nextOffset and incompletePage.nextOffset, showing awareness of the result shape. Given three modes, nested objects, and 14 parameters, it covers selection, invariants, and paging well; it says nothing about failure modes when a selector is ambiguous or unmatched, which is the remaining gap.

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 100%, so the schema already documents every parameter including the mode/view enums and nested genotype/geneResults shapes. The description reinforces the mode-to-selector coupling and the 'uniquely resolved' genotype constraint but introduces almost no syntax or format detail beyond what is already structured. Baseline 3 is appropriate.

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?

States a specific verb and resource ('Browse CPIC recommendations') and then partitions it into three named modes with the exact selector each requires. It also names sibling tools (pgx_search_pairs for drug identifiers, pgx_list_reference for phenotypes), so an agent can locate this tool in the family without opening schemas.

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?

Gives explicit selection logic per mode and for the view toggle ('view=test_alerts selects complete example CDS alerts instead of recommendations'). It routes the agent to pgx_list_reference for valid phenotype values. It stops short of stating when to prefer pgx_get_guideline or pgx_lookup_diplotype instead, so exclusions are incomplete.

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