execute_action

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

Annotations already signal openWorldHint and destructiveHint, but the description adds critical context about untrusted element names and the full return schema including refs_expired and error keys. This goes well beyond the structured annotations.

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?

The description is well-structured with a front-loaded purpose statement, a security warning, return value details, and a compact Args section. Every sentence adds value, and the formatting makes it easy to scan.

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 mutation-heavy tool with an output schema and error handling, the description covers return keys, the refs_expired workflow, and untrusted input risks. It is complete enough for an agent to invoke correctly without additional context.

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 0% schema description coverage, the description compensates by explaining ref's source, enumerating allowed actions ('click', 'hover', 'type', 'select', 'press_key'), and clarifying that value is used for type/select. However, the meaning of value for press_key is not fully specified.

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 opens with 'Execute an interaction on a page element by its ref number,' clearly stating the verb, resource, and key input. The list of action types further distinguishes it from siblings like get_page_map (which produces refs) and fill_form (form-specific).

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?

It instructs that ref comes from the Page Map Actions section and explicitly says to call get_page_map before retrying when refs_expired is true. It doesn't draw direct comparisons with fill_form or scroll_page, but the workflow is clear.

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