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

Annotations only supply readOnlyHint and openWorldHint=false, but the description goes well beyond: it explains needed_to_start_cents as a held balance that can exceed the estimate and is partly refunded, that scrape estimates are an upper bound versus usual_cost_cents, and that charge_note explains zero-cost cases. That is exactly the billing/auth context an agent cannot infer from structured fields.

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 core purpose is front-loaded, but the body is one dense paragraph of stacked parentheticals ('for a Flow too, where Flows on lists are on') that is hard to parse on first read. 'Uses the same calculation as the Sorzio add-on' adds little for an agent deciding how to call it.

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 17-parameter tool with no output schema, the description covers returns only for the Flow case and leaves the response shape for ai/verify/scrape/enrich partially implicit. It also never explains how the model, prompt, or reasoning-effort inputs move the price, which is the whole point of the tool.

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?

At 53% schema coverage the schema does part of the work, and the description adds real meaning for spreadsheet_id (returns output_columns and reused_columns) and list_id (list row counts, Flows-on-lists). However, it says nothing about how model, prompt, system_prompt, reasoning_effort, web_search, or the provider enums affect the estimate, and it silently drops the 'enrich' kind.

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: 'Estimate what an AI, verify, scrape or Flow run would cost', and enumerates the three input modes (row count, list, connected sheet). That distinguishes it from the run_* siblings, though it never explicitly says it does not execute anything. The description omits the 'enrich' kind that the schema's enum allows, so the stated scope is narrower than the actual scope.

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?

Usage is implied rather than stated: 'for your workspace the estimate is what Sorzio will charge' and 'Signed out, estimates are list prices. Free.' tell the agent what the numbers mean but never say when to prefer this over run_ai/run_flow or that it should be called pre-flight to check affordability. No exclusions or sibling routing are given.

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