snap_anomaly_check

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

Annotations already declare readOnlyHint and openWorldHint, and the parenthetical '(read-only)' merely restates that, earning no credit. Beyond annotations, the description discloses substantive behavior: the baseline is the average of days before the last two, today is excluded, day boundaries follow SNAP_REPORT_UTC_OFFSET, and low-spend days are not judged. It stops short of saying what the response looks like.

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?

Content is front-loaded in a question that captures the intent, followed by the mechanics and constraints. The anomaly enumeration is long but each item is informative; a little tightening of the parenthetical examples would improve it.

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 two-parameter read-only check with no output schema, the description covers what is compared, the baseline definition, the exclusion of today, timezone handling, and the low-spend guard. The main remaining gap is that it does not characterize the shape of the result (e.g., verdict vs. metric list), which matters since no output schema exists.

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 50%: min_daily_spend is documented in the schema, while baseline_days has only constraints (3-29, default 7) with no description. The narrative explains the baseline concept but not how baseline_days interacts with it, so it adds only modest meaning beyond the structured fields.

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?

The opening question plus 'Compares the last full local day with the usual day' names a specific operation and resource, and enumerates exactly which anomalies are surfaced (spend spike/drop, zero or falling purchases, ROAS drop, cost-per-purchase spike). 'Account level' and the broken-pixel/store example make it clearly distinguishable from sibling checks like snap_health_check or snap_billing_check.

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?

The framing question 'Did something change sharply yesterday?' implies the when-to-use case, and 'Account level' plus 'today is left out' set usage boundaries. However, it never names an alternative sibling check or states when NOT to use this (e.g., for historical or campaign-level investigations), leaving routing to inference.

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