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

Goes well beyond the destructive/openWorld annotations: it discloses that no personal data transits the call, that findings are written into `status_column` as found/no_match/failed, the precise no_match guard condition, that identical inputs share one lookup, that up to 8 rows run in parallel, and that a timeout yields `stopped: "time_budget"`. It also flags the excluded async steps (dropcontact, fullenrich, lemlist, pipecorn) and the hosted-agent `tool_not_in_job` / `missing_tools` path — none of which the annotations convey.

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 purpose is front-loaded, which is right, but the body is a dense wall of nested clauses and backticked fragments ('a row whose steps failed or never ran stays empty, its code in errors'). The final hosted-agent paragraph describes a narrow edge case at a length disproportionate to its frequency.

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 usefully documents the return shape (counts plus `errors: [{row_id, code}]`, `remaining`, `stopped`), the retry loop, and the dry_run plan — exactly the gaps a caller would otherwise hit. It falls short only on the long tail of tuning parameters, leaving an agent to guess at `providers`, `max_paid_steps`, and `qualify`.

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?

With 68% schema coverage the schema already carries much of the meaning, and the description mostly reinforces it (status_column lifecycle, dry_run). Many of the 22 parameters — providers ordering, qualify/qualify_min, technologies, max_paid_steps, people_per_row, batch_size, the _org/_group/_project scoping params — get no treatment in the description, so the coverage gap is not compensated.

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?

Opens with a specific verb and resource ('Run the email waterfall over a TABLE's rows, on the server') and pins the op to `update_rows`. It distinguishes itself from the single-contact path ('the one-contact tool collects them'), though it never names `waterfall_email` outright as the alternative, so the routing contrast 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 concrete operating rules: only rows with an empty `status_column` are taken, call again with the same arguments until `remaining` is 0, clear a status to retry a single row, `dry_run` plans without calling. It also explains the filter/row_ids exclusion. What's missing is a direct statement of when to prefer `waterfall_email` or `waterfall_person_act` instead.

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