Reference image

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

With only readOnlyHint:false and openWorldHint:false in annotations, the description carries the behavioral burden and delivers: importing places the image on a locked, semi-transparent layer above the art, size changes use nearest-neighbour resampling, and a photo scaled to 32×32 is explicitly characterized as a silhouette starting point. This discloses mutation, layer placement, transparency, and resampling behavior that annotations alone cannot convey. No contradiction with the annotations.

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?

Roughly 4 sentences, purpose-first: main goal, compact op list with embedded semantics, and a single closing caveat that earns its place by preventing misuse. No fluff or repetition of the schema. The op list is dense yet scannable.

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 tool with 9 parameters, 4 ops, and an output schema (so return values need not be described), the description covers the ope semantics, the laryal behavior, opacity context, and a size-related quality warning. Minor gaps remain: 'remove' and 'list' behaviors are left to the reader's inference, and the relationship between opacity and the import op is only implicit. This is solid but not exhaustive.

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 coverage is 56%, so the schema partially documents parameters (size, sprite have descriptions), and the description adds genuine meaning to the critical required parameter `op` by explaining what each enum value does behaviorally. It also aligns 'semi-transparent' with the opacity parameter. Other parameters like x/y are positional and self-evident with defaults, so the lack of dedicated parameter prose is acceptable, though a 5 would require systematically richer guidance for opacity and size.

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 specific verb+resource pair ('Bring an external image into the sprite as a reference layer') and gives the concrete use cases (tracing proportions, sampling colours). The four ops are each given a one-line semantic ('import' places a locked, semi-transparent layer; 'sample_palette' reads colours without importing), which clearly delineates the tool's scope from siblings like palette, layer, and read_pixels.

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?

The description conveys clear context for each op: import is for placing a reference layer, sample_palette is for reading dominant colours 'without importing', list and remove round out lifecycle management. It also gives a practical caveat about when an import result is acceptable (a starting point, 'never a finished sprite'). It does not explicitly name sibling alternatives or state when-not-to-use, so it stops short of a 5.

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