execute_extension_command

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

Annotations only provide destructiveHint=true; the description goes further by naming which kinds of commands are destructive ('closing apps, deleting files') and instructing the agent to confirm with the user before running those, which is actionable context beyond the flag. It does not cover permissions, side effects on driver state, or timing, leaving a small gap.

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?

Every sentence carries load: purpose/examples, prerequisite, routing rule, arg convention, return convention, destructive warning. It is somewhat dense and reads as a wall of clauses, but nothing is redundant and the most important scoping rule ('prefer a dedicated tool') is stated early.

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 generic dispatch tool with a free-form nested `args` object and no output schema, the description covers the key unknowns: how to discover valid command names, how to supply IDs, and how elements are returned. Minor gaps remain around error behavior for unknown commands and arg type coercion.

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 coverage is 100%, so the baseline is 3, but the description adds real semantics: element IDs from find_element should be passed as plain strings, usually under `elementId`, and result element references come back as plain ID strings. That clarifies the nested `args` object's expected key conventions beyond the schema's generic 'Named parameters' note.

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 ('Run') and resource ('any extension command ... by name') with two concrete examples of command naming ('windows: getWindows', 'pcomm: getScreenText'). It explicitly positions itself relative to siblings ('Prefer a dedicated tool when one exists'), so an agent can distinguish it from the dozens of dedicated element/window tools without opening a schema.

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?

Gives an explicit prerequisite ('Call list_extension_commands first to get the exact name and parameters') and a clear when-to-use/when-not rule ('Prefer a dedicated tool when one exists — use this for commands without one, including commands added by plugins'). It also adds a confirmation policy for a subset of commands.

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