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

Annotations already establish readOnly/idempotent/non-destructive, and the description adds genuinely new facts: the call is free and consumes no browser session, and it returns specific content categories (browser choice, sign-in link, method, screens, menus, actions, notes, dedicated tool pointers). That content inventory is the main value-add over annotations. It does not describe response size, caching, or staleness, which would be the next tier of disclosure.

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 in the first clause, but the sentence is long and list-choked, packing return contents, trigger condition, fallback behavior, cost, and a LiveChat action enumeration into two dense sentences. The trailing LiveChat action list is arguably redundant with the sibling livechat_* tools and dilutes the core instruction. It is informative but could be trimmed without losing meaning.

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 must convey what comes back, and it does enumerate the return contents in enough detail for an agent to know the guide will cover browser choice, sign-in, method steps, and app-specific tool pointers. Combined with the clear pre-browser-step trigger and the no-session/cost note, an agent has what it needs to call it correctly. Minor gap: nothing on behavior when the app is ambiguous or unknown.

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 defaults (app defaults to 'LiveChat (Text)'), so the schema already does the semantic work. The description hints that app names map to Cotlet slugs/catalog membership and implies the task shapes the returned method, but never explicitly explains what 'task' changes in the output. Baseline 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 makes clear this tool returns procedural guidance (which browser, sign-in link, step-by-step method, screens/menus/actions, notes, dedicated tools) rather than performing browser actions, which distinguishes it from the many cotlet_browser_* siblings. It also states coverage scope ('works for any web app; one outside Cotlet's catalog gets the general method') and even enumerates LiveChat actions. It falls short of a crisp single-sentence 'returns X for Y' statement, since the opening phrase describes the workflow rather than the return value.

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?

There is an explicit trigger: 'Call it before the first browser step in an app,' which tells the agent exactly when to spend a call. It adds useful qualifiers — free, no browser session — implying no cost to calling early and no session state required. It stops short of naming when-not to call or pointing to a specific alternative when the app is outside the catalog beyond the generic fallback statement.

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