diff

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

With no annotations, the description carries the full burden and does disclose at least one useful operational trait: a 90-second timeout. However, it does not explicitly state whether the operation is read-only, what pre-existing artifacts are required, or whether any device/state is modified during comparison.

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 description is three short sentences with no filler. The purpose is front-loaded, the mode guidance is immediately actionable, and the timeout is an extra worthwhile operational fact.

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 complex tool with 22 parameters and no annotations, this description is too thin. It does not explain which selectors/identifiers are needed to perform a meaningful diff, how the screenshot mode is requested when the schema only allows 'snapshot', or what operational prerequisites exist. The output schema covers return values, but the invocation context remains under-specified.

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?

For a tool with 22 parameters, the description adds very little parameter-level guidance and never explains how the two snapshots or screenshots to compare are selected. The mention of snapshot vs screenshot does not map cleanly to the schema because 'kind' only permits 'snapshot', and the many common selectors documented in the schema are not tied to any invocation pattern.

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 names a concrete action ('Compare... snapshots or screenshots') and an explicit outcome ('identify UI changes'), which distinguishes it from capture-oriented siblings like snapshot and screenshot. However, the advertised screenshot mode is not backed by the schema: the required 'kind' parameter is constrained to 'snapshot' only, so the description is slightly inconsistent with the input schema.

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 gives actionable mode-selection guidance: use snapshot comparisons for semantic tree changes and screenshot comparisons for pixel differences. It does not state when to prefer sibling tools or when not to diff, but the primary mode decision is clearly explained.

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