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

Annotations already declare readOnly/idempotent/closed-world, so the safety profile is covered. The description adds real behavioral context beyond them: the algorithm used, that each point reports before/after averages from 4 observations, and how many following observations stayed on the new level. It does not mention cost, result ordering beyond `limit`, or pagination behavior, keeping it short of a 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?

Front-loaded with the core definition, then the method, then the output shape, then usage routing — a sensible order with no filler prose. It is somewhat long and repeats the whole definition in Portuguese, which is serviceable for a bilingual product but costs density.

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 15 parameters and no output schema, the description carries the burden of explaining results, and it does so by enumerating the returned fields and the 4-observation before/after window. What it does not resolve is the window-length default discrepancy, which is the one thing an agent must get right to trust the returned dates.

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 the structural baseline is 3. The description does add interpretive value (weekly vs daily series, step magnitude meaning for `sensitivity`, change vs changePct in the output), but it also introduces a conflict: 'Default period 52 weeks' versus the schema's `days` default of 28. That ambiguity in the window parameter is a real gap for an agent choosing whether to set `days` explicitly.

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 and resource ('Finds the dates where clicks/impressions moved to a new sustained level'), names the underlying technique (robust CUSUM), and explicitly carves out what it is not ('Isolated spikes are not change points (see detect_anomalies)'). An agent can distinguish this from detect_anomalies or find_decaying_content without opening any 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?

Gives a concrete when-to-use case ('Use to date a lasting drop or gain and then correlate with get_annotations') and an explicit exclusion routing to a sibling. It also frames the typical causal scenarios (Google update, migration, content launch), which is exactly the selection signal an agent needs.

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