Get Page View

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

Annotations provide readOnlyHint=true and openWorldHint=true, and the description reinforces both consistently while adding substantial context beyond them. It discloses the token-budget guarantee, what the completeness block reports for closed shadow roots and cross-origin frames (dual counts rather than a single zero), and the exact fallback behavior of the `since` read. None of this is derivable from the annotations, and it correctly explains the open-world behavior of reporting what was NOT read and why.

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?

Despite its length, every sentence earns its place: the core purpose is front-loaded, followed by return contents, then parameter guidance, then edge-case behavior. There is zero filler or tautology. The density is justified by the tool's genuine complexity, and the structure flows logically from what it returns to how to control it to what its limits are.

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?

With no output schema, the description correctly explains the return payload (identity, landmark regions with expansion costs, interactive surface refs, digest/app skeleton, form/table inventories, unread account, and next call). It covers the complex edge cases of shadow roots and iframes thoroughly. The only incompleteness is the unexplained `view` and `detail` parameters, which leaves an agent guessing at part of the input surface. This is a small but real gap in an otherwise comprehensive definition.

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 0%, so the description carries the full parameter-documentation burden. It explains `location` (scopes to one region ref), `budget_tokens` (with a concrete subagent value), `mode` (links at real cost), and `since` (cheap repeat read). However, three of the seven parameters—the required `page`, plus `view` and `detail`—receive no explanation. While `page` is fairly self-evident from the name, `view` and `detail` are not obvious and their interaction with the orientation read is undocumented. This is a genuine gap given the zero schema coverage.

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 states a precise, specific purpose: 'Read a page as an ORIENTATION, not a transcript, under a token budget it never exceeds whatever the page size.' The verb-resource pair is unmistakable, and the 'not a transcript' contrast actively differentiates it from sibling tools like get_text. An agent immediately knows this is the high-level structural read, not a raw text extraction.

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 description embeds strong usage direction: `budget_tokens=2500` suits a subagent, `mode='links'` includes in-prose links, and `since=<read_token>` is the cheap repeat-read strategy with explicit fallback semantics. The 'not a transcript' phrasing plus the open/closed shadow root and same-origin/cross-origin iframe rules tell an agent exactly when this tool is appropriate versus alternatives. It even specifies when the since-read fails over to a full read, which is precisely the kind of decision guidance that matters.

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