Fill page field

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

Annotations already establish mutability and non-idempotency, so the description correctly carries no repetitive safety claims. The session/identity isolation disclosure adds genuine behavioral context beyond annotations: it tells the agent that the write is scoped to a session and that identity isolation requires separate sessions, which is operationally important for an openWorldHint=true tool. Slightly more detail (e.g., overwrite behavior on re-fill) would push it higher.

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?

Exactly two front-loaded sentences: the first delivers the core action with scope, the second adds the critical usage constraint. Zero filler words; every element serves purpose and usage communication.

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?

Given the tool's genuine complexity (nested iframe selectors, multiple locator strategies, 6 params, 0% schema coverage), two sentences feel thin. The description covers page and session addressing well, but omits any mention of the iframe-chain capability, behavior on locator timeout (timeoutMs exists), or what happens when the field can't be found. The presence of an output schema and annotations mitigates some burden, landing this at minimum viable.

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 coverage and a highly complex polymorphic locator parameter (role/text/label/placeholder/testId/css strategies, iframe chains up to 5 deep, exact-match flags), the description should compensate significantly. It only loosely maps to two of six params ('form field'→value/locator, 'explicitly addressed page'→pageId, 'session'→sessionId) while ignoring the locator complexity, ref-slot, and timeout semantics. The description's contribution is limited to the session-scoping nuance.

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 uses a specific verb+resource ('Fill a form field') with a clear scope qualifier ('located on one explicitly addressed page') that establishes a page-addressability constraint. It clearly differentiates the tool's scope from page-level operations, though it doesn't explicitly distinguish itself from close form-related siblings like browser_check or browser_select_option (e.g., it doesn't say 'for text inputs, use fill; for checkboxes, use check').

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?

The session-scoping rule ('value is entered only in that session') provides clear context for when this tool applies, and 'use separate sessions for different identities' is a genuine usage directive. However, it never explicitly names alternatives among the 35 siblings or states when NOT to use this tool in favor of browser_press, browser_check, or browser_select_option.

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