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

Annotations only declare read-only, open-world, non-destructive; the description adds genuinely new behavioral facts the agent could not infer: ai_overview is always null because the provider never returns it, and for some countries (e.g. 'tr') the provider returns organic-only so feature_data_available is false and nulls mean unknown, not absent. It even gives a remediation (retry with country="us").

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 and the trigger question are front-loaded in the first two sentences, then the caveats are isolated into a short bulleted list. Every sentence carries operational information — no restating of the title or filler — and the most decision-relevant caveat (null means unknown) is bolded in substance by placement.

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?

An output schema exists, so return-value shape need not be explained, and the description covers the two failure modes that would otherwise cause misreads of null fields. What is missing for a four-parameter audit tool is any statement about device/language behavior or cost/latency implications beyond the 25-query cap.

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 only 25% (only 'queries' is documented, including the max-25 cost bound), so the description must carry the rest. It partially does: it explains the consequence of the country parameter (the 'tr' organic-only case and the country="us" workaround). However, device and language semantics are never addressed, leaving two of four parameters with no guidance in either structured field or 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 first sentence states a specific verb and resource: fetch the SERP for a list of queries and report which click-eating features are present. It names the concrete features (Featured Snippet, Map Pack, People Also Ask) and explicitly distinguishes itself from find_ai_overview_suspects and ga4_ai_traffic, so an agent can separate it from sibling SERP/AI tools without opening a schema.

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 a concrete trigger scenario ('why is my CTR so low at position 3?') plus explicit routing away from this tool: use find_ai_overview_suspects for suspected AI Overview queries and ga4_ai_traffic for AI assistant traffic. That is when-to-use, when-not-to-use, and named alternatives in one compact block.

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