Set recording mode for message boxes

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

Annotations already declare the safety profile (readOnly/idempotent/non-destructive), yet the description adds substantial behavioural context: the auto-timer with autoCloseSeconds, that a human click pre-empts the timer, that concurrent tools/call are refused with 'dialog_open' plus remaining seconds, and that boxes are reported in _meta.dialogs[]. It also reconciles the state-change nature of the tool with readOnlyHint by explicitly stating 'No undo step, no change to the project' and resetting the mode after a server start.

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 purpose is front-loaded, but the body is a single run-on construction stitched together with dashes and semicolons ('...shows the real box on screen for screenshots, and a timer closes it after autoCloseSeconds...', '...refused with dialog_open and the remaining seconds'), which makes it hard to scan. Every fact is relevant, but the structure costs readability.

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 high-complexity session-mode tool with an output schema, the description covers purpose, side effects, lifecycle and interaction with other calls well, and it need not restate the return shape. The one material hole is the consumption semantics of the default `scope=next` (used up by the next mutating call even if no box appeared), which matters for correct invocation yet appears only in the schema.

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 all four parameters (including `scope`) are already documented, which sets the baseline at 3. The prose adds colour for `visible` and `autoAnswer=enter` (fixed dates move, result differs from a non-recorded call), but `scope` — arguably the most error-prone parameter — is never mentioned in the description, so it does not exceed the schema.

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 first clause gives a specific verb+resource ('Switches the recording mode of this session') and the remainder immediately operationalises it: subsequent message boxes are shown on screen for screenshots. No sibling in the list competes with it, and no sibling does anything similar, so the agent can identify it instantly.

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 states the operative context clearly: switch the mode on before a call that would raise a message box, run that call in the background, and take the picture meanwhile. The strategy of running the box-showing call in the background is exactly the guidance an agent needs, but there is no explicit 'when not to use' statement and the `scope` choice (next vs session) is left entirely to the schema.

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