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

The description far exceeds the annotations' disclosure. It reveals auth-required behavior ('if the MCP needs a credential/login, invoke returns a connect link'), payment edge cases ('if it is paid and the wallet is empty, invoke returns a checkout/top-up link'), permission requirements ('Writes... require workspace owner/admin'), and side effects (invoke runs 'without adding the MCP to the toolkit and without bloating the tool list'). No contradiction with annotations exists — description's write-oriented behavior is consistent with readOnlyHint: false.

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?

It's a wall of text with no bullet points or headers, yet nearly every sentence earns its place — there's almost no filler. The 'KEY:' marker provides a useful emphasis in an otherwise flat structure. The density is justified given the tool covers 14 actions and two distinct subsystems (MCP catalog + prompt library), but it would benefit from paragraph breaks or bullet lists for scannability.

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 23-parameter, 14-action dispatcher with no output schema, the description covers an impressive amount: the core flow, install-vs-invoke distinction, auth and payment edge cases, prompt library, permission model, and the role of every sibling-worthy action (report_bug, request_mcp, list_tools). Minor omissions: no error-handling guidance (e.g., invalid tool_id, network failures) or rate-limit expectations, and the installed_in_toolkit vs installed_in_workspace flags are mentioned but not explained. Still, this is near-complete coverage for an exceptionally complex tool.

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?

With 0% schema coverage and 23 parameters, the description carries the full semantic burden. It does explain the action enum thoroughly in prose (each of the 14 values gets contextual meaning), which is the most important parameter. However, concepts like arguments, prompt_vars, conversation, tool_id, and mcp_id are never explicitly mapped to the actions that need them — it's inferred. The prompt library parameter names (prompt_title, prompt_body, prompt_slug) are guessable but undocumented. The description earns a solid middle score for nailing the action parameter but leaving the remaining 19 parameters unexplained.

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 clear definition: 'The official mcp.ai marketplace — the in-platform catalog of every MCP/tool, AND the way to run them.' It uses specific verbs (search, describe, invoke, install) and explains the action-dispatch pattern. It differentiates from siblings by explicitly scoping what each action family does and names the distinguishing behavior 'KEY: invoke works even when the MCP is NOT installed' — the kind of detail that sets it apart.

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?

Usage guidance is exemplary: 'Use install only to make an MCP PERMANENT in the active toolkit... prefer invoke for a single/occasional use' provides an explicit when-to-use vs. when-to-avoid directive. It also names alternatives for sibling tools ('list_tools lists what is callable right now', 'request_mcp asks us to build a NEW MCP when nothing fits') and describes the decision flow: search → describe → invoke/install.

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