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

Annotations already declare read-only and non-destructive, and the description adds valuable behavioral detail: signal_source=stream rows are minutes-fresh, capped_since_utc encodes when a campaign hit 100%, and it returns avg-daily-spend and budgetUsagePercent polls. It also notes the data-source boundary (Marketing Stream) and where full history lives, giving the agent a solid behavioral model.

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 packs several important distinctions (same-day vs history, signal_source semantics, SP and budgetUsagePercent return types) into a moderately sized block. It is front-loaded with the core purpose and uses parentheticals to add detail without excessive length, though the abundance of acronyms slightly reduces skimmability.

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?

For a read-only alert tool with no output schema, the description covers what it returns (alerts, SP, budgetUsagePercent), the freshness semantics, and where to get full history. It omits pagination and date-format details, but those are largely inferable from the schema and the tool's read-only nature, so the remaining gaps are minor.

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 60% (end_date and start_date lack descriptions), and the description does not clarify those parameters or how period_days/threshold_pct interact with the alert logic. It mentions output concepts like signal_source and capped_since_utc, but those are return fields rather than parameter semantics, so it fails to compensate for the schema gap.

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?

Description opens with a specific verb+resource ('Find campaigns approaching or exceeding their daily budget'), immediately distinguishes same-day stream-based alerts from full-history ppc_query datasets, and enumerates additional returned alert types (SP, budgetUsagePercent). It clearly differentiates from siblings like ppc_campaigns and ppc_query.

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 provides explicit context for when this tool is relevant (SAME-DAY where Marketing Stream data flows) and points to ppc_query datasets for full history, effectively naming the alternative. It stops short of explicit when-not-to-use phrasing across all siblings, but the guidance is clear enough.

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