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

Annotations only provide high-level hints (readOnlyHint=false, destructiveHint=false, openWorldHint=true). The description adds critical behavioral details: invoke works even when MCP not installed, requires credentials or payment possibly, returns connect/checkout links, and writes require workspace owner/admin. It also explains the distinction between installed_in_toolkit vs installed_in_workspace. This goes beyond annotations and is very valuable. Missing a bit on idempotency or error cases, but good coverage.

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 single long paragraph that is information-dense but not structured (no bullet points or clear action mapping). It is somewhat lengthy and streams many details without clear organization, making it hard to parse. It could be improved by separating the marketplace flow from the prompt library and using bullets for each action. It is not wasteful but could be better structured.

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?

Given the enormous complexity (actions, prompt library, install vs invoke, auth), the description covers many critical points like auth requirements, payment, and the difference between one-off and permanent installs. However, it lacks explanations for several actions (resume, subscribe, cancel details) and does not describe return formats for commands. Considering the huge schema (23 params), the description is not fully complete to guide an agent without more 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?

The schema has 0% description coverage, with 23 parameters but the description does not map the parameters to the actions. It mentions 'action=search', 'tool_id', 'arguments', 'mcp_id', and 'prompt_slug' implicitly, but most parameters (e.g., 'request_name', 'prompt_title', 'conversation', 'immediate', 'tier_slug') are left entirely to the schema without clue as to which action uses them. Given the low coverage, the description must compensate but only does minimally.

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 identifies the tool as the marketplace for MCPs and prompts, with a detailed breakdown of actions (search, describe, invoke, install, etc.). However, the scope is very broad (encompassing both MCP catalog and prompt library), which slightly muddies a single clear purpose. It does distinguish from siblings by positioning itself as the main platform hub versus specific tools like sefaz_ro_nfce_completa_consultar.

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?

The description provides substantial guidance on when to use each action: 'search discovers MCPs by intent → describe returns one MCP's full profile... invoke RUNS that tool', and explicitly contrasts invoke vs install ('prefer invoke for a single/occasional use; use install only to make an MCP PERMANENT'). It also mentions when to use list_tools and subscribe/cancel. However, it does not explicitly state when NOT to use this tool versus alternatives outside the sibling set, and some actions like resume are not explained, lowering the score.

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