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

Annotations already mark the tool as read-only, idempotent, and non-destructive. The description adds valuable behavioral context beyond those annotations: it is 'deterministic', it is a 'paid resource' with a specific price, and it returns a 'technical failure category and evidence basis'—critical because no output schema exists. It does not cover rate limits or edge cases, but the annotation-safety profile lowers the bar for those.

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?

Two dense sentences cover purpose, inputs, output, exclusions, and pricing with no filler. The most decision-relevant facts—classification scope and return type—are front-loaded. The only minor redundancy is 'Paid' appearing twice, but it does not hurt effectiveness.

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?

Despite high complexity (9 undocumented parameters, no output schema), the description omits any detail about most input signals and does not explain how the returned evidence basis is structured. An agent could understand the tool's purpose but cannot confidently construct a correct request or interpret the full response from this definition alone.

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?

The schema has 9 parameters with 0% description coverage and no enums, so the description must compensate. It only broadly references 'HTTP status, error text, protocol and execution/payment signals', leaving booleans like permission_ok, payment_signature_present, output_match, and schema_ok completely unexplained. This is insufficient for an agent to know what values or combinations are valid.

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 names a specific verb and resource: 'paid deterministic classification of agent-commerce failures' from supplied signals. It further clarifies output as 'technical failure category and evidence basis' and explicitly excludes 'legal or fraud adjudication', making the tool's purpose distinct from legal/fraud classifiers even without naming a sibling.

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 clearly describes the intended inputs ('HTTP status, error text, protocol and execution/payment signals') and provides an explicit exclusion: 'no legal or fraud adjudication', serving as a when-not-to-use. However, it does not name alternative sibling tools or give precise trigger conditions for choosing this tool over related failure-analysis tools.

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