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

Annotations already declare readOnly, idempotent, non-destructive, and not open-world, so the safety profile is covered. The description adds two genuinely non-derivable facts: 'Free, no browser session' tells the agent this does not open or consume the Cloud Browser, and the content list discloses what the response contains. It does not say anything about output size or whether the guide is cached, but the important behavioral caveat is present.

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 comes first and is followed by the timing instruction, so it is front-loaded correctly. The long comma-delimited enumeration of covered content is somewhat run-on and could be shortened, but every element listed is decision-relevant rather than filler.

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 by enumerating method, screens, menus, actions, notes, and app-specific tools. For a two-parameter, read-only lookup tool that is close to sufficient; only the returned format or size is left unspecified.

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%, with both `app` and `task` documented including the PT Distinction default, so the baseline of 3 applies. The description adds context about app naming ('any web app', catalog vs. non-catalog) and enumerates the PT Distinction actions supported, which helps the agent frame the `task` argument. It does not explain how much detail `task` should carry or what happens if `task` is omitted.

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 outcome: it returns a playbook for operating a given web app through Cotlet's browser tools, covering browser choice, sign-in link, step-by-step method, screens/menus/actions, and any app-dedicated tools. That clearly separates it from the cotlet_browser_* siblings, which act rather than instruct. The phrasing 'How to work in an app' is a bit abstract on its own, so it leans on the enumerated content list to become concrete.

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 an explicit trigger point, which is exactly the routing signal an agent needs before falling back to browser_click/browser_navigate. It also notes 'Works for any web app; one outside Cotlet's catalog gets the general method,' pre-empting the question of unsupported apps. It stops short of naming a when-not case or explicitly pointing at the dedicated pt_distinction_* tools as an alternative.

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