ps_selection

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

The description adds real behavioral detail beyond the annotations: 'subject' invokes AI Select Subject, 'mask' adds a layer mask derived from the selection, and feather/expand/contract/smooth take a numeric value. That said, it advertises a 'delete' (and 'fill') step that overwrites pixel content irreversibly, while annotations declare destructiveHint=false — a real inconsistency for a mutation tool.

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?

One dense but well-ordered sentence: action menu first, modifiers with their value slots, then the optional post-selection operations. No filler, and the most important information (what actions exist) is front-loaded.

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 7-parameter, single-action dispatcher with no output schema, the description covers nearly all actions and explains the mask/fill/delete outcome. Remaining gaps are minor: copyToLayer is unmentioned, ordering when multiple modifiers are combined is unspecified, and return values/selection state after the call are not described.

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?

With only 29% schema description coverage, the description carries most of the semantic load: it gives the rect/ellipse coordinate order [x,y,w,h] that the schema leaves as an unlabeled array, ties 'value' to the feather/expand/contract/smooth actions, and reveals that 'color' is a hex fill color. It omits the 'then' enum value 'copyToLayer', which appears in the schema but not in the prose.

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 specific resource (Photoshop selections) and enumerates every supported action verb — all, none, invert, rect, ellipse, subject, sky, layerTransparency, feather, expand, contract, smooth — which maps directly to the action enum. It is clearly a selection-manipulation tool and distinguishable from siblings like ps_layerOps or ps_applyFilter, though it never explicitly says so.

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?

Usage is implied by the action list and the 'then' clause, so an agent can infer the intended flow (make a selection, then optionally fill/delete/mask). However, there is no explicit when-to-use guidance versus alternatives such as ps_removeBackground, ps_contentAwareFill, or ps_layerOps, and no prerequisites (e.g., an open document, an active layer).

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