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

Annotations already cover the safety profile (readOnlyHint, idempotentHint, destructiveHint=false), yet the description adds substantial non-obvious behavior: credit cost of 1 per record with a 1-credit minimum, free timeouts, the async 'fetch_queued' state, CO offset cap of 1000 and 5000-row/day quota, and the obligation to retain the CO 'attribution' field. These are exactly the operational constraints annotations cannot express.

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 purpose and the required parameter, and nearly every clause carries unique operational information. It loses a point because the zip and bounding-box filters are described after the return-field list rather than grouped with the other filters, making the sentence order harder to scan than the content warrants.

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 12-parameter tool with no output schema, the description covers return fields (parcel_id, site_address, lat/lon as WGS84 centroid, attribution), cost model, pagination, async recovery, and per-state quotas. An agent could call this correctly and handle the queued case without consulting anything else; only 'detail' is unaccounted for.

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 0% schema description coverage, the description carries the full burden and mostly succeeds: it defines state as a 2-letter code, county as an exact name with sample values, city and address as prefixes with an example format, zip as 5-digit, the bbox as all-four-or-nothing with a 0.5-degree cap, and limit 1-100 with offset paging. The clear gap is the 'detail' parameter, which is never mentioned anywhere and defaults to 'basic'.

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+resource ("Search US property parcels") and immediately narrows scope with the required 'state' parameter and the explicit exclusion "never owner data." An agent knows exactly what domain this touches; the only missing piece is an explicit contrast with near-neighbors like query_permits or site_screen, so it stops short of a 5.

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 when-to-use routing: "If the area is not loaded yet, returns status 'fetch_queued' with a job_id: poll get_fetch_job, then search again" — a full recoverable-workflow instruction naming the sibling to call. It also states the Colorado-specific prerequisite (county/city/address filter required) and how to page with offset, so nothing about invocation order 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.