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

Annotations only declare readOnlyHint=false, openWorldHint=true, idempotentHint=true, destructiveHint=false; the description adds the operationally critical facts the agent cannot infer — that it returns immediately, that records begin landing within a minute, that a full run takes minutes, and that returned receipts are already real records.

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 return behavior is front-loaded with 'RETURNS AT ONCE', and each subsequent sentence carries distinct information: background execution, polling cadence and stop condition, receipt validity, and idempotent re-calls. No sentence restates the schema or annotations.

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 no output schema and only two fully documented parameters, the description covers the remaining unknowns an agent needs: that results are not returned synchronously, how to observe progress, when to stop, and that duplicate calls are safe. Nothing required to invoke it correctly is missing.

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 description coverage is 100% (subject = product name, topic, or URL; terms = optional concepts), so the schema already carries the parameter meaning. The description reinforces 'subject' and 'product or topic' but adds no syntax, format, or terms guidance beyond the schema, so the baseline of 3 applies.

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 description states a specific verb and resource ('Start collecting evidence on a product or topic from public sources') and immediately identifies its role as the starter of an asynchronous harvest, which is exactly what separates it from retrieval siblings like search_evidence and category_warmth.

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 gives an explicit follow-up procedure: poll search_evidence or category_warmth after ~30 seconds, keep calling while counts grow, stop when they stop. It also resolves the obvious ambiguity about repeat invocation by stating that a second call for the same subject joins the existing run rather than starting a new one.

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