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

Annotations already cover the safety profile, and the description adds substantial context beyond them: it spends workspace balance, it mutates the target column (consistent with readOnlyHint=false), it returns a run_id immediately and continues in the background, and it documents the idempotency contract where reusing a request_id returns the same run without a second charge. It also includes a prompt-injection warning about row values. Minor gap: it doesn't say whether results are appended as a new column versus in-place for sheets, beyond the overwrite hint in the schema.

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?

Purpose and the provider rule are front-loaded, and nearly every sentence carries a distinct operational rule. The one structural weakness is the placement of 'Matching is exact and case-sensitive' immediately after the output-value enumeration, which reads as if it describes the result values rather than the only_rows_where filter, and the dense paragraph could be broken into steps.

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 nine parameters, nested target objects, a confirmation handshake, and no output schema, the description still covers what an agent needs: cost checking, the request_id contract, the confirmation loop, test runs, and how to poll for the result via get_run.

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 a baseline of 3 is warranted, but the description genuinely adds cross-parameter relationships: list runs use output_name while sheet runs use target.sheet.output_column, request_id must be fresh per run but reused on retry, and confirm_token/test_confirm_token are only valid after a needs_confirmation result. That is meaning the schema alone doesn't convey.

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 sentence states a specific verb and resource ('Check whether each email in a column is deliverable'), and the surrounding text makes clear how this differs from siblings like run_ai, run_scrape, and run_flow by naming the exact deliverability outcomes written back.

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 explicit routing rules: call estimate_cost first, omit provider rather than asking the user, use test_rows to try a few rows, and poll get_run with wait_seconds until the run ends. It also spells out the when-not path via needs_confirmation (nothing started until the user says yes).

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