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

Annotations already cover the safety profile (readOnlyHint, idempotentHint, openWorldHint), so the bar is lower. The description adds real behavioral context beyond them: the before/after proxy semantics, that 'regions' returns per-rectangle stats without a delta, and that preview output is gated by privacy.send_previews_to_llm. It still says nothing about cost or size of the proxy computation.

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 sentences with the core purpose front-loaded before the mode breakdown, and no filler. The second sentence is dense with parenthetical asides, but each clause carries information the schema alone doesn't convey, so nothing is wasted.

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 an 8-parameter tool with an output schema (so return values need not be explained here), the description covers both modes, their applicable parameters, and the preview privacy dependency. The only minor gap is timing/cost expectations for proxy rendering, which is not critical for correct invocation.

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 100%, so the schema already documents mode semantics, mode-scoping of region/region_a/region_b/include_previews, and defaults. The description restates mode semantics in prose and clarifies the before/after meaning, but adds little syntax or constraint detail beyond what the schema fields already say. Baseline 3 is appropriate.

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?

States a specific verb and resource ('Compare before and after, or two regions') and then defines the two modes precisely, including what 'before' and 'after' mean (photo without filters vs with them). It does not, however, distinguish itself from nearby read tools such as get_histogram, inspect, or preview, so an agent gets the tool's own scope but no sibling differentiation.

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 mode breakdown effectively tells the agent which variant to pick: 'before_after' for measuring filter effects, 'regions' for two rectangles on the same filtered image with the explicit instruction 'no delta (compare them yourself)'. It gives no exclusions or pointers to alternative tools, but the context for each mode is clear.

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