Recommend the next advertising action

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

Annotations already declare readOnlyHint=true, destructiveHint=false and openWorldHint=false, so the closing 'Read-only' is redundant. However, the description adds real behavioral context beyond the annotations: the diagnosis taxonomy it applies, the signals it consumes, and the fallback behavior ('hold' with 'insufficient_data') instead of a forced recommendation. That fallback is the kind of trait an agent cannot infer from structured fields.

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-loads the purpose in the first sentence, then layers the failure taxonomy, the usage trigger, and the fallback rule in a logical order. The 'call vs. post office' analogy consumes a clause that is illustrative rather than strictly load-bearing, and the trailing 'Read-only' duplicates the annotation, so it is dense but not maximally tight.

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, the description need not document return fields, and it correctly only hints at the shape ('action hold', 'diagnosis insufficient_data'). Given 8 parameters at full schema coverage plus annotations, nearly everything an agent needs to call this correctly is present; only explicit sibling routing is absent.

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% across all 8 parameters, so the schema already defines sku, current_cpl, window_*, cpl_trend_days, competitor_active and the creative signals. The description only gestures at them collectively ('measured funnel, its current cost per lead, and optional competitor and creative signals') without adding format, units, or interaction semantics. 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 ('Diagnose why a product is underperforming and return the matching remedy') and enumerates the exact failure modes it distinguishes (structural loss, contested auction, traffic quality, offer or price mismatch, creative fatigue, weak offer). An agent can tell this apart from measure_sku_funnel or audit_ad_verdict, which measure or judge rather than prescribe a remedy.

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 an explicit trigger ('Use when a campaign misses its target and the cause is not yet established') and a justification for the tool's existence (opposite remedies for leads that fail on the call vs. at the post office). It also states a boundary condition — returns 'hold'/'insufficient_data' rather than guessing on small samples. It stops short of naming sibling tools as alternatives, so it is clear context without explicit routing.

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