eval

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

The description adds a key behavioral nuance beyond the `openWorldHint` annotation: the JS runs in an isolated world so `window.X = ...` mutations are NOT visible to page JS. This is exactly the practical behavior an agent must know before calling eval. It also mentions that `eval_handle` is the alternative to retain references, which is actionable and not deductible from the annotation alone.

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 a single sentence and front-loaded with the action, while still adding the critical isolated-world caveat and forwarding to `eval_handle`. It is a little dense with the semicolon-clause, but every piece earns its place. Nothing is fluffy or redundant.

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 tool with four parameters and a strong schema, the description covers the most important evaluation-context trap. However, it does not state what the caller should expect as a return value or how async expressions are handled, which agents often need to know when using an eval tool; leaves some usage ambiguity.

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?

The input schema has 100% description coverage for all four parameters, so the description does not need to repeat parameter meanings. The description adds no extra intel about `expr`, `lease`, `session`, or `timeout_ms`, so it does not improve on what the schema now gives; the baseline of 3 applies.

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 specific verb and resource: it 'evaluate[s] a JS expression in the page.' It further clarifies the scope by noting the isolated context and the boundary on `window.X` mutations, and it differentiates itself from `eval_handle` by telling the caller when to use that sibling instead. An agent can then distinguish `eval` from the many other page-tools.

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 explicitly provides an alternative: 'use eval_handle to retain references', which tells the agent when `eval` is insufficient. It also implies this tool is for cases where references are NOT retained. However, it does not define broader conditions (e.g., when it is better than `wait_fn` or extracting text) or list exclusions beyond the reference case.

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