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

Annotations already declare destructiveHint=true, readOnlyHint=false, openWorldHint=true, and idempotentHint=false, so the safety profile is covered structurally. The description adds no behavioral context beyond that: it does not say a new image is generated, that the operation is non-idempotent and consumes generation resources, or what happens to the source image. The input listing it does provide is parameter information, not behavior.

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?

Two tight sentences with zero filler, and the routing condition is placed first where the agent needs it. Nothing is redundant, though the brevity borders on under-specification for a five-parameter generative tool with nested image objects.

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?

An output schema exists, so return values need not be explained, but for a destructive, non-idempotent generative tool with five parameters, nested union types, and duplicate image channels, the description is thin. It omits how the primary-image fields and reference-image fields relate to each other, and gives no sense of what the fallback mode is capable of versus the named modes.

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?

Schema description coverage is 80%, so the baseline is 3. The description restates the parameter shape ('one main image, up to six reference images, and natural-language instructions'), matching the schema rather than adding to it. It notably leaves the biggest ambiguity unaddressed: why both `image` and `image_url` (and both `reference_images` and `reference_image_urls`) exist and which to prefer.

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 states a specific purpose: a catch-all workspace 'Custom Prompt' mode that applies natural-language instructions to a room image. It explicitly differentiates itself from siblings by declaring it applies 'when another named RoomLab mode does not match'. It stops short of saying what the tool produces (an edited room render), so it is clear but not fully self-contained.

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?

Routing guidance is explicit and front-loaded: use this only as a fallback when no named RoomLab mode fits, which effectively tells the agent when not to use it. It does not address edge cases such as what to do when a request partially matches a named mode or whether multiple concerns should be split across sibling calls.

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