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

Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds genuine value by describing the return payload (funnel step data for rendering plus a compact summary), which matters because no output schema exists, and by specifying how filters must be applied globally rather than per step.

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 is front-loaded and nothing is pure filler, but the middle sentence is a dense run-on ("A global filter or breakdown on a property or argument needs that column, with the same dataType, on every step's event; put a condition...") that is hard to parse on a single read. Breaking it into shorter directives would improve usability.

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?

The tool has a deep, nested four-parameter schema with no output schema, and the description only partially compensates: it covers filters, breakdowns, and return shape, but says nothing about the time window, metrics array constraints (min 2 steps, display.name required), or conversionCriteria. It leans on the funnel-queries skill for those details, which is reasonable but leaves the description itself incomplete for a high-complexity tool.

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?

With 75% schema coverage the schema already documents most parameters, but the description contributes semantics the schema does not: that a global filter or breakdown on a property requires the same column and dataType on every step's event, and that all-step conditions belong in global filters exactly once. That is meaningful guidance for constructing a valid spec.

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?

"Build and execute an ad-hoc funnel analytics query" gives a specific verb and resource, and "ad-hoc" implicitly separates it from create_funnel_query. However, it never names the sibling tools it replaces or differs from, so an agent must infer that distinction.

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 states real construction rules — put cross-step conditions in global filters once rather than per step, keep dataType consistent across steps, and use cohort filters/breakdowns to segment by behavior. It also makes loading the funnel-queries skill a prerequisite. What it lacks is any explicit comparison to create_funnel_query, update_funnel_query, or execute_query.

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