nex_browser_captcha_submit

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

Goes well past the annotations (idempotentHint=false, destructiveHint=true) by disclosing the actual cost model ("Charges the balance at the current server price"), the non-idempotency rationale (each Grid round has its own prepareId), server-side revalidation of product/team/page/Grid image, retry caps, and an uncertainty-handling procedure. This is exactly the extra context destructive/expensive tools need.

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 the primary action and follow-up tool, then layered with guards (no resubmit after timeout, attempt cap, no final button click, billing). Dense and nearly waste-free, though the billing/reconciliation clauses are slightly compressed and could be ordered more logically relative to each other.

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?

Covers everything an agent needs for a costly, destructive, non-idempotent 1-parameter call: precondition (prepareId from a prepare step), what gets charged, what is not done (final button), retry limits, and what to do when the result is ambiguous. No output schema exists, and the description acknowledges that ambiguity explicitly rather than pretending the result is deterministic.

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 only one parameter at 0% schema coverage, the description carries the burden and does add real meaning: prepareId comes from a prepare step, must be reused when calling complete, and is per-Grid-round. It stops short of stating where to obtain it or any format/lifetime rules, so it is strong but not complete.

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 ("Submit a prepared page CAPTCHA once") and immediately pairs it with the follow-up sibling action (nex_browser_captcha_complete with the same prepareId), drawing a clean boundary against both nex_browser_captcha_prepare and nex_browser_captcha_complete. It also carves out scope it explicitly does not cover ("Does not click the final sign-up/login button").

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 when-to-use ("call it directly when the user's task needs it"), an explicit counter-case ("a price inquiry is not a purchase request"), and hard negatives ("Never resubmit after a timeout", "At most 3 attempts per resource flow", reconcile before any retry). The agent has a complete routing and retry policy without needing to inspect siblings.

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