Recalibrate CPL bounds

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

Annotations already declare readOnlyHint=true, and the description confirms 'Read-only.' Parity there. The description adds real behavior: it falls back to measuring the product itself when rates are omitted, and it reports drift vs. the current baseline. It does not disclose cost/latency of on-demand measurement, keeping it out of the 5.

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, each load-bearing: what is computed, the source substitution, the fallback, the trigger, and the flag. Verb+resource is front-loaded, and 'Read-only' is correctly placed last as a closing guarantee.

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 explanations are unnecessary. For a 6-param read-only tool the description covers purpose, fallback behavior, and the business flag adequately. It omits when observed_* should be supplied vs. left null (beyond 'when omitted'), a minor 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 each parameter is documented in the schema. The description adds only the fallback semantics ('measured on demand when omitted'), which is already stated per-param in the schema. It does not explain the observed_* vs. window_* interaction beyond that. Baseline 3 applies.

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 (recompute) and two named resources (Stop CPL, Goal CPL), defines each parenthetically in business terms, and distinguishes the computation source ('from its own measured funnel rates instead of portfolio-wide assumptions'). An agent can separate this from siblings like measure_sku_funnel.

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?

Explicit trigger: 'Use whenever a cost-per-lead target is being set, defended, or questioned.' Names the specific alert condition it surfaces (assumed target above true break-even, invisible in the ads dashboard). No sibling is named as an alternative, but the when-to-use clause is unambiguous.

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