An MCP server for an e-commerce plant shopping application with product catalog, shopping cart, and Gr4vy payment processing integration. Provides interactive widgets for browsing plants, managing cart, and checkout.
Three tools with complete schemas and descriptions, but significant gaps in parameter documentation and error handling. Tool names follow verb_noun convention (list_, show_, start_). Descriptions are present (100-200 chars) and explain purpose, but lack actionable guidance for LLM selection. Parameters have types and enums where applicable, but descriptions are sparse or generic. No documented error handling, recovery paths, or output schemas. The 'show_cart' and 'start_checkout' tools reference UI widgets that handle display, which is good separation of concerns, but the tool descriptions don't explain what data structure is returned or how the LLM should interpret the response.
List available Plantly plants and show the interactive catalog. Supports filtering by plant care attributes: sunlight needs, watering frequency, and climate/hardiness zone. The UI widget handles all display — do NOT repeat product names, prices, or descriptions in your response. Just present the widget silently.
Show the shopping cart with the given items. The UI widget handles all display — do NOT repeat item names, quantities, or totals in your response. Just present the widget silently.
Start the checkout process. Shows order summary and Gr4vy Embed for payment. This should be called when the user wants to pay for items in their cart. The user will enter payment details via the Gr4vy Embed widget. The UI widget handles all display — do NOT repeat order details or payment info in your response. Just present the widget silently.
No documented output schemas. Tools return data to render UI widgets, but LLM doesn't know what fields/structure to expect. This breaks downstream tool chaining and forces LLMs to guess.
Parameter descriptions are minimal or missing context. 'search' param in list_products says 'Free-text search for semantic matching' but doesn't explain what fields are searched or expected format. 'items' array in show_cart/start_checkout lacks description of what each item object represents.
No error handling or recovery guidance. Tools don't document what happens on invalid input (e.g., invalid enum values, malformed items array). No guidance for LLM on retryable vs fatal errors.
Inferred effective spec: 2025-06-18+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 65 | 2025-06-18+ | v2 |
start_checkout is marked WRITE but lacks confirmation/dry-run pattern. Checkout initiates payment, a destructive operation that should support confirmation before execution to prevent accidental charges.
Tool descriptions instruct LLM not to repeat data ('do NOT repeat product names, prices, or descriptions in your response'). This is a UI concern, not a tool contract. Descriptions should focus on WHAT the tool does, not HOW the LLM should format output.