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

Beyond the readOnly/openWorld annotations, the description discloses substantive behavior: named-company queries are refused, small cohorts are suppressed, and responses carry confidence and suppressed_below_k metadata. It also sets an interpretive boundary that results are historical aggregates rather than live performance.

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?

Four sentences, front-loaded with purpose before return shape and caveats. Dense and mostly waste-free, though the caveat sentence could be tightened slightly.

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 needn't explain return fields — yet it still names the key outputs (recipes, columns/rows, confidence, suppressed_below_k) and covers refusal and suppression constraints. Nothing critical for a correct call is missing.

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 every parameter is already documented in the schema. The description echoes the filter dimensions (channel, format, audience type, bid strategy) but adds no syntax, allowed values, or semantics beyond what the schema provides, and omits num_results entirely — 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?

The description states a specific verb and resource — search aggregate historical campaign configurations — and enumerates the filter dimensions (channel, format, audience type, bid strategy). This distinguishes it from siblings like search_experiments, though it never names those alternatives explicitly.

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 gives useful context about what the results represent ('historical configurations, not guaranteed outcomes or the current account's performance') and when queries are refused, which implicitly steers usage. However, it never says when to prefer this tool over a sibling like search_experiments or experiment_performance_stats, leaving the choice to inference.

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