watch

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

Annotations already declare idempotentHint=true and non-destructive, so the description doesn't repeat those. It adds valuable context beyond annotations: 'Survives extension idling, not a browser restart' clarifies persistence scope, and 'Events are collected, not pushed' explains the event delivery model. These details help an agent understand lifecycle and consumption without contradicting the 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 compact, front-loaded with the core purpose, and includes illustrative examples. It packs multiple behavioral notes (polling, event collection, persistence) into three sentences without redundancy. Slightly dense but efficient; no wasted words.

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 an 11-parameter tool with no output schema, the description explains the watch mechanism and lifecycle but does not describe what the tool returns (e.g., event structure or format). Since parameters are fully documented in the schema, the main gap is the output contract. An agent might need additional clues to interpret the result, especially for the poll action.

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 100%, so each parameter is individually documented. The tool description does not add additional meaning to the parameters beyond what the schema provides; it only references interval_s in context. This meets the baseline for fully covered schemas, but does not elevate it.

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 the tool's purpose: to keep watching a page in the background for conditions like text appearing, disappearing, or changing. It provides concrete examples (pipeline green, Approve button appears) and explicitly distinguishes the polling mechanism ('events are collected, not pushed'). This differentiates it from siblings like watch_dom or wait_for by emphasizing the background, interval-based monitoring.

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 gives usage context (polling with action=poll, blocking script with --wait) but does not explicitly state when to choose this tool over alternatives like watch_dom or wait_for. It implies usage for long-running background checks but lacks exclusions or alternative routing, leaving the decision partially to inference.

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