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

Annotations already establish the safe, idempotent, read-only profile, and the description adds genuinely new behavioral facts: it costs nothing and opens no browser session ('Free, no browser session'), and it degrades gracefully to a general method for uncatalogued apps. The only soft spot is the phrase about 'clicking in the user's Cloud Browser', which could momentarily read as if this tool performs the clicks; 'no browser session' corrects it. No contradiction with readOnlyHint or openWorldHint, since the tool itself reaches nothing external.

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 core instruction an agent needs ('Call it before the first browser step in an app') is buried in the third clause rather than front-loaded. The opening sentence is a long colon-list that mixes purpose with returned content, and the trailing fragments ('Works for any web app...', 'Free, no browser session.') read as afterthoughts. All content is relevant, but the ordering costs scanability.

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 carries the burden of saying what comes back, and it does so concretely enough (browser choice, sign-in link, method, UI map, notes, companion tools). For a zero-required-parameter lookup tool the remaining gaps are minor: nothing on output size or whether results are cacheable/persistent across steps. An agent knows when to call it and roughly what it will receive.

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 coverage is 100% and both parameters carry their own descriptions, including the default app. The description adds only indirect color: 'which browser to use for this workspace' hints that the app argument resolves against workspace context, and the uncatalogued-app note explains what happens when the app isn't recognized. It does not explain the task argument's effect on output, so it stays at the baseline for well-documented schemas.

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 enumerates the payload (browser choice, sign-in link, step-by-step method, screens/menus/actions, notes, dedicated tools), which tells an agent this is a pre-flight guidance lookup rather than an action tool. It implicitly separates itself from the cotlet_browser_* siblings by framing itself as the thing you call before the first browser step. The opening phrase 'by clicking in the user's Cloud Browser' is slightly misleading about whether this tool itself clicks, but the enumerated contents resolve that by the end of the sentence.

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?

'Call it before the first browser step in an app' gives explicit timing/ordering guidance that an agent can act on. 'Works for any web app; one outside Cotlet's catalog gets the general method' sets coverage expectations. It stops short of naming alternatives (e.g. cotlet_apps for discovering what is in the catalog), so the boundary with that sibling must be inferred.

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