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

Annotations provide only generic flags (readOnlyHint=false, openWorldHint=true), so the description carries the transparency burden, and it delivers richly: detailed one-off execution behavior ('runs it pontualmente (one-off), without adding the MCP to the toolkit and without bloating the tool list'), auth flows (invoke returns a connect link when credentials needed, checkout/top-up link when wallet empty), write-permission requirements, and prompt-library share behavior (opens without login). No contradiction with annotations — in fact it deepens what readOnlyHint=false implies by specifying exactly which operations mutate.

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 description is a dense wall of text covering 14 actions across two domains. It front-loads the core concept and uses 'Core flow:' and 'KEY:' markers to highlight critical info, which helps navigation. However, given the bulk and the two distinct sub-domains (MCP operations vs prompt library), it would benefit from more structured separation. The length is somewhat justified by the complexity, but readability suffers. Every sentence earns its place, yet structure is only average.

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?

Despite the high complexity (14 actions, 23 params, no output schema, generic annotations), the description is remarkably thorough: full end-to-end flow, edge cases (credentials, empty wallet, retry), permission model, and the second prompt-library domain fully explained. Weaknesses: it never specifies per-action return formats (which matters with no output schema) and action-specific nuances for params like immediate, tier_slug, and conversation are absent. Strong but with identifiable gaps.

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 0% for all 23 parameters, so the description must compensate. It does explain the semantic meaning of several key params through workflow context (action driving the whole flow, tool_id/arguments for invoke, mcp_id for describe, query for search). However, many parameters remain unexplained in the description (prompt_title, cancel_comment, conversation, immediate, tier_slug, request_name, report_context, etc.), leaving real gaps for the agent. Partial compensation, not full.

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 clearly states the tool is the official mcp.ai marketplace catalog plus the runner for MCPs, with a specific scope. It explains the core action flow (search→describe→invoke) and distinguishes itself from siblings like authenticate and show_version by covering catalog/execution/billing/prompt-library functions. It's broad because it covers 14 actions across two domains (MCP marketplace + prompt library), which slightly dilutes the single-verb clarity, but the purpose is unambiguous.

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?

Extremely explicit guidance on when to use each action: 'prefer invoke for a single/occasional use' vs 'Use install only to make an MCP PERMANENT', 'Search/describe flag installed_in_toolkit vs installed_in_workspace', and clear alternatives like 'request_mcp asks us to build a NEW MCP when nothing fits'. The sequential core flow (search→describe→invoke) provides step-by-step triggering context, and permission requirements are stated (writes require workspace owner/admin). This is exemplary usage guidance.

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