Runnable MCP server demonstrating a free tool plus a paid tool gated by @harpd/mcp-paid-tool-starter. Works with Claude Desktop and any MCP client.
Two tools with complete input schemas and descriptions. Naming follows verb_noun convention (get_weather, premium_forecast). Descriptions are adequate (50-100 chars) but lack actionable context for LLM selection. The premium_forecast tool includes a complex payment object parameter with good structure, but descriptions do not explain prerequisites, error recovery, or when to use each tool. Output schemas are not documented. Error handling is minimal, no recovery guidance or actionable error messages. The payment parameter structure is well-designed but the tool descriptions do not explain the payment flow or failure modes.
Current weather for a city. Free.
24-hour premium weather forecast with confidence intervals. Costs $0.01 per call.
Output schemas not documented. LLMs cannot plan downstream calls or extract structured data without knowing response field names and types.
Error handling lacks recovery guidance. Errors return isError=true with text but do not tell the LLM what to do next (retry, ask user, or fail). Example: 'Unknown city' error should suggest 'Try one of: tokyo, san-francisco, new-york'.
Tool descriptions lack context for LLM selection. 'Current weather for a city. Free.' does not explain when to use get_weather vs premium_forecast, or what 'free' means in the context of a paid-tool server. Descriptions should state prerequisites and differentiate from similar tools.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | 2026-07-28+ | v2 |
premium_forecast payment parameter lacks clear error guidance. If payment proof is invalid or missing, the error should explain: 'Payment required. Provide a valid x402 receipt with amountUsd >= $0.01 and a valid txHash starting with 0x.'
No dry-run or confirmation step for premium_forecast. Paid tools should support a preview or confirmation before charging. Current design risks accidental charges if the LLM misunderstands the payment flow.