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

Annotations only state openWorldHint=true with no readOnly/destructive hint. The description expands meaningfully: it reveals that invoke has a hidden one-off install behavior, explains the connect/checkout link side-effects, discloses the 'owner/admin' permission requirement for writes, and reveals that search/describe flag installed_in_toolkit vs installed_in_workspace. It explains the billing lifecycle via subscribe/cancel. No contradiction with the sparse annotations — the openWorldHint=true is consistent with a tool that operates on the outside world of marketplaces and profiles.

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?

This is a dense, run-on paragraph that buries critical usage decisions mid-sentence. The key guidance ('prefer invoke for a single/occasional use') could be a bullet point instead of being woven into the middle of a cascade of commas and arrows. The description jumps from marketplace flow to invoke internals to billing to permissions to prompt-library details without markdown, headers, or line breaks. While every sentence earns its place, the lack of any formatting for a 3-domain tool makes it mentally taxing to parse.

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 tool of this complexity — 14 actions, 23 params, multiple side effects (installations, billing, prompt publishing) — the description covers the primary flows comprehensively: search→describe→invoke, install, subscribe/cancel, report_bug, request_mcp, and all three prompt actions. It also covers the auth edge-cases and one-off execution. The few gaps are the 'resume' action, which is never mentioned (unclear if it belongs to subscribe or the prompt flow), and vague params like 'message' or 'conversation', which remain undocumented. The lack of an output schema means the reader only learns return shapes for a few actions (get_prompt, publish_prompt).

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 free-form parameters (no descriptions), the burden is on the description. It directly explains actions like 'search,' the difference between describe/invoke semantics, and the prompt-actions' return values: 'get_prompt returns its full text with {{variables}} filled, and publish_prompt saves a prompt and returns a shareable mcp.ai/p/<slug> link.' It covers tool_id and arguments ('every tool with its id + params'). However, it never touches on ~1/3 of parameters (cancel_reason, report_context, conversation, resume action), which remains a gap given the schema offers zero hints on its own. So close to a 5 but not exhaustive.

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 statement that this is the marketplace catalog and runner of MCPs, then uses specific verb+resource pairs for every capability: 'action=search discovers MCPs by intent → describe returns one MCP's full profile → invoke RUNS that tool,' plus 'search_prompts finds one, get_prompt returns its full text.' It clearly differentiates from siblings by explicitly calling out that 'report_bug sends feedback' and explaining how this catalog relates to permanent toolkit installation. The description of the catalog's scope (MCPs AND a prompt library) leaves no ambiguity about its dual purpose.

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?

Explicitly says when to invoke vs install: 'prefer invoke for a single/occasional use' and 'Use install only to make an MCP PERMANENT in the active toolkit.' It gives decision support: 'invoke works even when the MCP is NOT installed,' 'If the MCP needs a credential/login, invoke returns a connect link; if it is paid and the wallet is empty, invoke returns a checkout/top-up link.' It also distinguishes the prompt library as being 'about ready-made prompt TEXT rather than MCPs' and specifies that 'Writes... require workspace owner/admin' — clear user-role and auth statements.

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