Act on the page

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

Annotations only flag openWorldHint and destructiveHint, but the description adds substantial behavior: return shape (recorded steps + new snapshot), verified-selector generation with a `fragile: true` marker for positional fallbacks, the refusal condition while a run item awaits verification, and hard safety rules (only the user decides what to submit, upload only named files, never type passwords). This is far more than the annotations convey.

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 core action and return value, then layered with selector, safety, and gating details. It is dense with parentheticals but nearly every clause carries operational weight, so it reads as compact rather than padded.

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 destructive, open-world tool with no output schema, the description supplies the return shape, the selector-quality signal, the safety constraints, and the refusal condition an agent needs before invoking. Coverage is strong; only minor gaps remain around what happens on partial batch failure.

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?

Top-level schema coverage is only 33% (actions and snapshot are undocumented), but the description compensates by explaining the `inspect` mode's behavior and the batch semantics of the actions array. The nested action fields are well described in the schema itself, so the description adds meaning where the schema is thin without needing to restate the rest.

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 (performs actions), resource (elements of the current page), and mechanism (by ref), plus the side effect of recording a task and returning steps with a fresh snapshot. An agent can distinguish this from browser_snapshot, browser_open, and the run_* tools without opening any 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 clear routing context: `inspect` acts on nothing and is named as the input for step_repair, batching multiple steps in one call is endorsed, and the tool is 'Refused while a submitted run item awaits verification.' It stops short of explicit when-not-to-use exclusions against siblings like browser_open, but the operational context is strong.

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