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

Annotations (readOnlyHint=false, openWorldHint=true, destructiveHint=false, idempotentHint=false) already signal this is a mutating, non-idempotent tool. The description adds meaningful behavioral context beyond those hints: writes require workspace owner/admin, invoke performs a hidden one-off install without bloating the toolkit, and auth/checkout gateways produce retry flows. Not quite a 5 because exact error/edge behavior isn't disclosed, but it substantially enriches the annotation baseline.

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 dense run-on paragraph mixing two domains (marketplace MCP handling and the prompt library) without any structural separation. While information-dense, it is hard to scan and the em-dash-heavy prose is taxing to parse. Front-loaded content helps, but the lack of paragraph breaks or bullet structure hurts readability for a tool with this much scope.

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 tool's complexity — 23 parameters, no output schema, 14 action variants spanning marketplace and prompt library — the description is remarkably complete. It covers the core flow, permission model, install-vs-invoke distinction, retry semantics after connect/checkout, and the prompt library operations. Minor gaps (individual parameter semantics, expected return structures) exist, but for such a broad multi-purpose tool, comprehension is high.

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%, so the description must compensate, and it does meaningfully. It explains the pivotal `action` enum values (search/describe/install/invoke/list_tools/subscribe/cancel/report_bug/request_mcp and the three prompt actions) and their purposes, plus how mcp_id/tool_id/arguments fit the flow. With 23 parameters it can't document every one, but the high-leverage control parameters are contextualized. A 5 would require covering more of the 23 params (e.g., limit, conversation, prompt_vars), so 4 is appropriate.

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 specifies the exact purpose: it is the mcp.ai marketplace catalog and execution engine, covering capability requests and the prompt library. It clearly states the core flow (search → describe → invoke) with specific verbs and resources, and no sibling tool (authenticate, connect, sefaz_pb_nfce_consultar, etc.) overlaps with this capability. This exceeds mere purpose clarity and firmly distinguishes the tool.

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?

Explicit when-to-use guidance is provided: 'prefer invoke for a single/occasional use' vs 'use install only to make an MCP PERMANENT'; subscribe/cancel for billing, report_bug for feedback, request_mcp when nothing fits. The description also calls out exclusions (invoke works even when MCP not installed, returned connect/checkout links require user action then retry). This is textbook usage guidance with named alternatives.

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