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

The description adds important behavioral context beyond the annotations: it explains that invoke works even when the MCP is not installed and runs one-off without bloating the toolkit, it mentions connect/checkout flows return links rather than errors, and clarifies that write operations require workspace owner/admin. While the annotations are minimal (not read-only, not destructive), the description proactively discloses these nuances. It doesn't describe every action's side effects (e.g., uninstall behavior), but for a multi-action tool this is substantial. No contradiction with annotations, so no score penalty.

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 dense but well-organized, front-loading the overall purpose and then flowing from core discovery flow to nuanced execution semantics (invoke vs install), permissions, and finally the prompt library extension. Every sentence adds distinct value; there is no filler. However, it is long and may benefit from bullet points to ease scanning, but still earns a high score for its efficiency.

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 high complexity (14 actions, 23 params, no output schema), the description covers the main workflows, distinguishes invoke from install, mentions auth/payment edge cases, and outlines the prompt library side. It does not enumerate every action's exact response format or error handling, but it gives enough operational guidance to make the tool usable. A 5 would be reserved for something like per-action parameter substitution, but this is well above average.

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 23 parameters with 0% coverage (no parameter descriptions), but the overall description explains the core parameters (action with its enum values, mcp_id, tool_id, arguments, prompt-related params) by embedding them in the workflow explanation. It clarifies the meaning of action values like invoke, install, search_prompts, and mentions prompt variables. However, not every parameter is explicitly explained (e.g., immediate, tier_slug, conversation, prompt_targets), so it doesn't fully compensate for the missing schema coverage, but it does add meaningful semantics that the agent cannot infer from the bare JSON.

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 that marketplace is the official mcp.ai catalog and execution platform, explicitly listing all its capabilities (search, describe, invoke, install, etc.) and distinguishing it from a simple tool. It names the target resource (MCPs and prompts) and the actions performed, making it unmistakable even among sibling tools like authenticate or connect.

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 explicit guidance on when to use which action and how they relate: 'Core flow: action=search... → describe... → invoke RUNS that tool', contrasts invoke vs install ('Use install only to make an MCP PERMANENT... prefer invoke for a single/occasional use'), mentions auth/wallet edge cases ('invoke returns a connect link', 'returns a checkout/top-up link'), and tells users to use list_tools for currently callable tools and request_mcp when nothing fits. It also notes write operations require owner/admin. This exceeds expectations.

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