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

Annotations already declare destructiveHint=true, readOnlyHint=false and idempotentHint=false, and the description goes well beyond them: exact cost model (1 credit per billable address, cross-referenced to total.billable in neverbounce_get_job), irreversibility ('cannot be stopped or undone'), a required safety parameter, and a 1000-credit ceiling for this tool.

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?

Front-loaded with 'WRITE' and the cost warning, then the safety rule, then the precondition, then the free-sample caveat, then the endpoint. Every clause carries decision-relevant information with no filler.

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, the description still tells the agent what financial effect the call has and where to check the billable count beforehand. For a destructive, credit-spending operation with 3 parameters, nothing material 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 coverage is 100%, so the baseline is 3; the description still adds value by tying max_credits to a concrete enforcement rule and by pointing to neverbounce_get_job's total.billable as the figure it is compared against, and by clarifying that run_sample spends nothing. It doesn't add new syntax or edge cases for job_id.

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 ('STARTS a created job'), its billing consequence, and the underlying endpoint. It is clearly distinguished from neverbounce_create_job, which makes the job but does not run it.

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 explicit preconditions ('must be in status waiting or waiting_analyzed'), an explicit exclusion with an alternative ('start bigger jobs in the NeverBounce dashboard'), and a free escape hatch ('run_sample true ... spends nothing'). Nothing about when to call it is left to inference.

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