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

This is where the description earns its keep: 'opens nothing by itself' and 'returns how to do it step by step' tell the agent the tool is purely advisory, which is behavior the readOnly/idempotent annotations imply but don't spell out. It adds the hand-off target (cotlet_browser_*) that annotations cannot express. It stops short of auth requirements or what the step payload actually contains.

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 operative content (guide-only, then use cotlet_browser_*) is front-loaded and clear, but 'from ChatGPT, Claude or any AI agent' and the closing claim about AnyList having no public API or official MCP server are promotional/meta filler that doesn't help an agent select or invoke the tool.

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 parameterless, output-schema-less guide tool, the essentials are present: what it returns, that it performs no action, and which tools to use next. Missing is enough detail on the return payload and on how the specific item/account context flows into the browser steps, which is the main practical uncertainty.

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?

Zero parameters gives a baseline of 4, but the total absence of inputs is itself semantically significant and unexplained: the agent has no idea whether the item is inferred from session state, named later in the browser flow, or must be passed via a follow-up call. That unanswered question is exactly what the description should resolve.

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 opens with a specific verb and resource ('Cross off item in the user's own AnyList account') and immediately clarifies the actual mechanic: it returns step-by-step instructions rather than performing the action. That distinguishes it from siblings like anylist_add_list_item, though the name/behavior mismatch (an action-sounding name for an advisory tool) costs it the top score.

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 does give a real workflow: call this to get the steps, then execute them with the cotlet_browser_* tools in the AnyList Cloud Browser. But it never says when to prefer this over cotlet_app_guide, and with zero parameters it gives no guidance on how the target item is identified, which leaves the agent guessing about the core selection problem.

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