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

Annotations only say read-only/idempotent; the description adds ordering, the earliestDay history boundary (before which no answer is possible), a truncated flag for result overflow, and the explicit limitation that the actor behind a change is never returned. It also explains the semantics of summary.key, summary.alert and the clientType taxonomy, which the agent cannot get from structured fields.

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 answer ('what was changed... newest first') followed by field semantics and then the date-span rule, so nothing important is buried. It is a dense run-on block with heavy parenthetical nesting, which costs readability slightly, but nearly every clause carries information an agent needs.

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?

There is no output schema, so the description carries the return-value burden and does so thoroughly: response field names (summary.from/summary.to, changePct, summary.key, summary.alert, clientType, earliestDay, truncated) are all explained. Combined with the date-span constraints, an agent has enough to call and interpret it correctly.

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 baseline is 3, but the description adds real meaning: from and to are inclusive, must be paired, cannot be in the future, and must be populated whenever the operator supplied a concrete month/week/date range. The default-30-day fallback behavior is only stated in prose, not in the schema.

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 ('what was changed in the ad and email accounts') across named platforms and enumerates the change classes covered (budgets, status, bidding, targeting, creatives, conversion tracking, sends). The ordering ('newest first') and the sibling distinction from configuration tools like get_campaigns (this is history, not current state) are clear.

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 explicit trigger questions ('what changed', 'why did spend or ROAS move on day X', 'did someone pause a campaign') and a strong exclusion-style rule: a named month/week or two dates MUST be sent as from/to, otherwise the tool silently reads the last 30 days. It does not name a competing sibling tool to route to instead, so it stops short of full alternative guidance.

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