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

Goes well beyond the annotations: it states the side effect (declined), the absence of notification to the applicant, irreversibility ('Cannot be undone'), and audit logging, plus the dry-run/confirm safety protocol. These are the exact behaviors an agent must know before a destructive call and are not derivable from destructiveHint=true alone.

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?

Three compact sentences with no filler, and the highest-stakes facts (decline, no notification, irreversible) are front-loaded before the dry-run mechanics.

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 destructive, idempotency-free mutation there is an output schema and full parameter coverage, so the description only had to cover consequences, reversibility, audit, and the confirm workflow — all of which are present.

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 already 100%, so the baseline is 3; the description earns an extra point by clarifying the default (dry run) and the exact expected value of confirm (the application's id), tightening what the schema expresses more abstractly as 'this change's exact target'.

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 ('mark an application as spam') and immediately discloses the concrete effect ('it is declined, and the applicant is not told'), which an agent can distinguish from approve_application or reject_application. It stops short of naming the alternative sibling (reject_application) or the boundary condition that would pick one over the other, so differentiation is inferable rather than explicit.

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 a clear precondition ('an open application') and the operational procedure (dry run by default; the real run needs confirm set to the application's id), which is exactly the context needed to invoke it. It does not say when not to use it or point to reject_application as the non-spam alternative, so it falls short of full routing guidance.

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