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

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds two genuinely useful facts beyond that: 'Free, no browser session' tells the agent the call costs nothing and does not open or consume a Cloud Browser session, which is non-obvious given the surrounding browser tools. It stops there, though — no indication of output size, whether guidance is cached or fresh, or what happens if the app is unknown beyond 'gets the general method'.

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?

Four sentences, all short except the first, which front-loads the core concept before enumerating contents. The list of what the guide covers is dense but each item is informative rather than filler. The only redundancy is the near-repeat of 'the app's screens, menus and actions' alongside 'the method step by step', which could be tightened.

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 only two optional parameters, rich annotations, and no output schema, the description carries the right load: it explains what the agent will get back and when to call it, so nothing essential for correct invocation is missing. The remaining gap is the effect of the task parameter and any expectation about response length or freshness, which matters for a guide tool the agent will read into context.

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%, with both parameters documented and a stated default (Kustomer), so the baseline is 3. The description adds a little meaning: it implies the app parameter drives catalog lookup ('one outside Cotlet's catalog gets the general method'). However, it never connects the 'task' parameter to output behavior — an agent cannot tell whether passing the user's task changes the returned method or dedicated-tool list, and 'in their words' is already in 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 description states the resource precisely (a per-app playbook covering which browser to use, sign-in link, step-by-step method, screens/menus/actions, notes, and the app's dedicated tools) and implicitly distinguishes itself from the many cotlet_browser_* action tools by being the thing you consult first. The opening phrase 'How to work in an app by clicking in the user's Cloud Browser' is slightly convoluted, and the description never uses a clear verb like 'returns' or 'retrieves', so the agent has to infer that this is a knowledge-lookup call rather than an action.

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' is an explicit, actionable when-to-use rule that positions it ahead of cotlet_browser_click/navigate/etc. 'Works for any web app; one outside Cotlet's catalog gets the general method' removes the main reason an agent would hesitate to call it for an unknown app. What's missing is guidance on repeat invocation (per app? per task? once per session) and any explicit exclusion, so it stops just short of a 5.

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