desktop_ui

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

Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds useful behavioral traits beyond that: compact ref emission, snapshot delta diffing, pattern filtering, and text observation. It omits, however, anything about snapshot lifetime/id validity, cost or latency of large traversals, and the shape of the returned tree — notable given there is no output schema.

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?

A single front-loaded sentence with the core verb+resource first and modifiers parenthesized. Dense with jargon but every clause maps to a real capability; a slight cost is that the parenthetical bundle is harder to scan than a short bulleted structure would be.

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 12-parameter, nested-schema tool with no output schema and no documented return structure, the description covers the conceptual capability but not enough operational detail — traversal limits, defaults, ref stability, and how the delta result is keyed are all absent. It is adequate to know what the tool is for, but thin for invoking it correctly on the first attempt.

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 only 8% (only `delta_of` is documented in the schema), so the description must carry meaning for 12 params including the nested `filter` and `text_target` objects. It partially compensates by naming compact refs, pattern filtering, delta diffing, and text observation, but leaves `hwnd`, `prune`, `max_depth`, `max_nodes`, `max_visited`, `skip_empty`, `text_budget`, and the five filter fields entirely unexplained.

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: 'Inspect a window's UI Automation accessibility tree', and names the downstream consumer of the `[s#.#]` refs (`desktop_ui_action`). This clearly separates it from action/click/type siblings, though it never explicitly distinguishes itself from `desktop_screenshot` (visual capture) as the alternative inspection path.

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 only implied by the mention of control refs 'for `desktop_ui_action`', suggesting an inspect-then-act workflow. There is no explicit when-to-use statement, no guidance on choosing between this and `desktop_screenshot`/`desktop_windows`, and no note on when delta mode or filtering should be preferred.

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