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

Annotations only declare readOnly=false, destructive=true, idempotent=false, openWorld=true, which tells an agent almost nothing about real consequences. The description supplies the useful mitigating fact that selection alone does not open or delete files, plus the staleness constraint implied by "fresh snapshot" and the "verified" qualifier suggesting target resolution can fail. It could go further on what happens to targets that no longer resolve.

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 tight sentences, purpose first then constraint, with no restatement of the title or schema. Nothing could be removed without losing information.

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?

The tool is complex (18 params, 7 target modes, open-world, destructive-flagged) yet the description is two sentences; an output schema exists so return values need not be explained, and the schema is fully documented, which lowers the burden. Still, target-resolution behavior and failure modes for stale or non-matching targets are left entirely to inference.

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% and every one of the 18 parameters is already documented in-schema, so the baseline applies. The description only gestures at the `targets` array ("identified targets") and adds no extra semantics about how id/label/name/role are combined for matching or how the top-level target enum interacts with per-target fields.

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?

"Select multiple verified native desktop items as one semantic accessibility operation" gives a specific verb (select), an explicit cardinality (multiple), and the mechanism (native accessibility), which distinguishes it from single-item computer_select/ui_select. It stops short of naming those siblings directly, so differentiation is inferred rather than stated.

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?

"Supply identified targets from a fresh snapshot" is a real precondition (element ids/labels must come from ui_snapshot/ui_find). However, there is no guidance on when to use this versus computer_add_to_selection, computer_select, or the ui_* equivalents, which is the main routing ambiguity among the ~15 selection-related siblings.

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