analysis_delivery_collapse_check

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

With no annotations provided, the description carries the full transparency burden and delivers: it declares 'Read-only', explains the same-weekday median baseline to prevent weekend false positives, states the current partial day is always excluded, notes it never reads action_log, and cites STRATEGY.md guardrail thresholds with defaults. This goes well beyond a typical one-line tool description.

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 description is long but information-dense and front-loaded with the core action. Every clause earns its place (baseline weekday logic, action_log independence, current-day exclusion, threshold defaults, read-only), and the final 'Read-only' is a clear safety capstone.

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?

Given the algorithm's complexity and the absence of an output schema, the description thoroughly covers input requirements, interpretation rules, and configuration sources. The one gap is the return shape — it does not describe what a result/signal looks like — but the detection semantics, platform-agnostic intent, and edge-case handling are complete enough for correct invocation.

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% with rich per-field descriptions, so the baseline is 3. The tool description adds beyond-schema value by explaining the ~30+ day row requirement, baseline algorithm, threshold parameter names/defaults, and that the current day is excluded — all helpful context for choosing and populating rows/as_of. It does not duplicate the schema's per-field semantics.

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 'Detect delivery collapse: campaigns whose impressions fell off a cliff while their status still says they should be serving' gives a specific verb, resource, and precise definition. It explicitly separates itself from google_ads_cost_increase_investigate ('inverse of'), making its purpose unmistakable.

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 description clearly states when to use it ('scheduled detector /daily-check runs'), what input to feed ('day-grain delivery report... for ANY platform'), and names a related tool it contrasts with. It does not explicitly exclude alternative analyze/diagnose siblings or state when NOT to use it, so usage direction is strong but not exhaustive.

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