Desktop view and images / 桌面显示与图像

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

Annotations only say readOnlyHint=false, destructiveHint=true, openWorldHint=false, idempotentHint=false; the description goes well beyond them. It discloses that the two image operations write PNG bytes to file_path on the MCP host and OVERWRITE that path (justifying the destructive hint), that return_image=false yields a saved-path message, that a failed preset still captures the current view and returns a warning instead of image content, that no solve runs and no prior view is restored, and that defaults are assignments rather than toggles.

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?

Front-loaded with the two core capabilities, then organized into an operation-routing block and a cross-cutting constraints block, which is appropriate for a ten-branch discriminator schema. However, it is dense and repeats some branch-level schema facts (set_unit defaults, display_ids omission behavior, screenshot overwrite), so a little trimming would help.

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?

No output schema exists, and the description compensates by explaining return shapes: confirmations for display operations, inline MCP Image content versus a saved-path message, and the text warning when a preset fails. Combined with write-path/overwrite behavior, authentication/connection requirements, and the no-solve/no-restore caveats, an agent has everything needed to call this correctly.

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 baseline is 3 and the per-operation field docs already do the heavy lifting. The description adds cross-cutting semantics not obvious from any single branch: defaults are assignments not toggles, display_ids applies BOTH flags and an omitted flag hides that label family, set_unit sends both units every time, and view_angle='current' preserves the camera. These are useful, but several restate what the nested $defs already say, so it does not fully clear the bar for a 5.

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?

Opens with a specific verb+resource+scope: control the desktop view and capture PNG images of the model or existing analysis results. It then enumerates all ten operations with a one-line role for each (camera/shading/labels/visibility, stage selection, screenshot, result plotting, units), so an agent can tell what the tool covers without reading the schema. It also names siblings it is NOT (query_results for numerical values, edit_geometry/edit_construction for entity/schedule edits).

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?

Explicit routing guidance: 'Use query_results for numerical values' versus plot_analysis_result for contours, and 'edit_geometry and edit_construction change entities and schedules' versus display-only operations. Prerequisites are stated (a connected desktop view is required) and the when-to-use split between save_model_screenshot and plot_analysis_result is clear. Nothing is left to inference.

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