MCP server that exposes the Bring! shopping list API to MCP clients such as Claude Desktop or Cursor
Server has 6 tools with complete input schemas and descriptions, but quality is uneven. Tool names follow verb_noun convention (get_lists, add_item, remove_item, complete_item, batch_update). Descriptions are present but brief (10-60 chars), below the 50-200 char baseline for LLM optimization. Parameter descriptions exist but lack detail on constraints, formats, and valid ranges. No output schemas documented. Error handling is minimal, no recovery guidance or categorization. batch_update has a complex nested schema but lacks clarity on operation semantics. Missing tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk levels assigned.
Add an item to a shopping list.
Apply a batch operation to multiple items.
Mark an item as completed.
Return a single Bring! shopping list with its items.
List all Bring! shopping lists.
Remove an item from a shopping list.
Tool descriptions are too brief (10-60 chars vs 50-200 char baseline). LLMs cannot determine when to select tools or distinguish similar ones. E.g., 'List all Bring! shopping lists' lacks context on when to call it vs get_list.
Parameter descriptions lack constraint details. 'item_id' is described as 'Item name, for example Milk' but does not specify format, length limits, or character restrictions. LLMs cannot validate input without explicit constraints.
No output schemas documented. Tool responses are unstructured, LLMs cannot plan downstream calls or extract required fields. E.g., get_list returns items but no schema shows field names, types, or pagination structure.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 64 | 2026-07-28+ | v2 |
batch_update operation parameter lacks enum constraint. Description says 'ADD, COMPLETE, or REMOVE' but schema does not declare enum values. LLMs may hallucinate invalid operations like 'UPDATE' or 'DELETE'.
No error handling guidance. Tools do not document what errors are retryable, user-fixable, or fatal. No recovery hints (e.g., 'If list not found, call get_lists() first'). Agents cannot self-correct on failures.