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

Beyond the readOnly/non-destructive annotations, the description discloses the most decision-relevant behavior: the fitting is done by a deterministic rule engine rather than the AI, unspecified zoning values silently fall back to typical values, and the run yields georeferenced IFC4, a UTM DXF site plan, and floor objects stored in Wai. It also bounds the result's authority ('pre-design screening, not an architectural design or a zoning approval'). These are exactly the traits an agent needs to frame and caveat the output.

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?

It is front-loaded with the core purpose and packs outputs, input format, defaults, and caveats into a compact block with no filler sentences. The long comma-chained sentence listing outputs is dense but each clause carries real information, and only the final reporting instruction feels like process guidance rather than tool definition.

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 13-parameter, no-output-schema tool, the description covers what an agent needs: what goes in (parcel geometry or point-plus-size, zoning defaults), what comes out (storeys, GFA/NFA, programmes, parking, and the specific validity checks), and where artifacts land (IFC4, DXF, floors in Wai). This compensates for the low schema coverage and the missing output schema; only the floors/gfa_m2 interaction is left ambiguous.

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?

Only 23% of the 13 parameters carry schema descriptions, so the description must compensate, and it does partially: it explains the two parcel input modes (corner-point list versus lat/lng plus parcel_l_m/parcel_w_m) and establishes that far_max, coverage_max, height_max_m and setback_m default to typical zoning values when omitted. However, several parameters remain unexplained in both places, notably how floors, gfa_m2 and name interact and what units or precedence apply when both gfa_m2 and floors are given.

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 description opens with a concrete verb and resource: it fits an office building or a residential block on a real parcel, and enumerates exactly what is computed (setbacks, footprint, TAKS, KAKS/FAR, height, storeys, GFA/NFA, workplaces/flats, parking, checks). An agent can immediately tell this is a parcel massing/feasibility calculator that emits building geometry, not a drawing or lookup tool. It stops short of a 5 because the nearest confusion risk (wai_design_house) is never named, so the boundary between house and office/residential-block design is only implied.

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 clear preconditions and input conventions: supply the parcel as lat/lng corner points or a single point plus parcel size, and note that zoning values default to typical values unless given ('say so'). It also sets an explicit behavioral expectation to answer with returned numbers rather than a pre-call estimate. The main gap is that no alternative tool is named and there is no explicit 'do not use when' condition beyond the screening-versus-approval caveat.

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