browser_snapshot

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

With no annotations, the description carries full transparency burden. It discloses key behaviors: returns structured elements with ref/box, waits for a DOM silent window, and marks offscreen elements in full mode. It does not explicitly state whether it is read-only or if it modifies page state, but the emphasis on snapshot and non-mutating language suggests safety. It could further explain determinism guarantees or side effects, but overall is reasonably transparent.

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 three sentences, tightly packed with essential information. It front-loads the core purpose, then provides behavioral details and mode distinctions without any fluff or redundancy. Every sentence contributes value.

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?

Despite having no annotations and no output schema, the description does not fully compensate. It explains the return concept (structured list with ref/box) but provides no details about the structure, pagination, or error behavior. The presence of many parameters that remain undocumented makes the tool's full behavior unclear, leaving the agent to guess on important configuration options.

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 must compensate. It only explains the 'mode' parameter (reading/interactive vs full) and mentions offscreen marking. It fails to explain the other six parameters (diff, scope, task_id, wait_stable, include_generic, include_offscreen), leaving the agent to infer their meanings from names and defaults. This is a significant gap for a tool with 7 configurable options.

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 clearly states it gets the page accessibility tree snapshot and returns a structured element list with ref/box. It uses a specific verb (获取/get) and identifies the resource (accessibility tree), distinguishing it from sibling tools like browser_screenshot or browser_evaluate.

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 provides usage context by explaining that reading/interactive modes only cover the viewport while mode=full includes the entire page, guiding when to choose each mode. It also mentions the event-driven deterministic snapshot that waits for DOM silence, which implies appropriate usage after page interactions. However, it does not explicitly compare against alternatives like browser_read or browser_wait_stable.

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