Loading ToolBench
A cross-platform e-commerce inventory synchronization engine that syncs inventory across multiple channels (Shopify, Amazon, WooCommerce), generates low-stock alerts, and automatically creates purchase orders
This MCP server has a single tool with critical definition gaps. The tool name 'sync_inventory' lacks a leading action verb (should be 'sync_inventory' → acceptable, but description is vague on WHAT gets synchronized WHERE). The tool description is moderately detailed (98 chars, within 10-1024 range) but lacks key operational details: does it actually sync data or just report differences? Is it idempotent? What happens on partial failures? Input schema shows two parameters with basic type definitions (array, integer) and descriptions, but 'channels_config' lacks an enum constraint, it accepts free-form strings like 'Shopify', 'Amazon_US', 'WooCommerce', which invites hallucinated channel names. No output schema is documented in the source, the tool definition in client.py returns a dict with specific fields (synced_skus_count, low_stock_alerts, auto_po_generated), but this structure is inferred from the example usage, not formally declared. The response structure is reasonable (includes counts and alerts) but lacks pagination guidance for the alerts array. No error handling is visible, no recovery guidance, no input validation, no categorization of failure modes. The tool modifies state (generates purchase orders, creates alerts) but this is not explicitly flagged in the description. Overall, this server would not pass code review: it lacks rigor in schema definition, missing formal output documentation, no error handling, and parameter constraints that should be enums are left as free-form strings.
Synchronizes inventory across multiple e-commerce channels, generates low-stock alerts for SKUs below threshold, and optionally creates automatic purchase orders
No output schema documented. The tool definition shows the method returns a dict, but the structure is inferred from example_usage.py, not formally declared in the tool metadata. LLMs cannot reliably extract fields or plan downstream calls without a documented schema.
'channels_config' parameter accepts free-form strings (Shopify, Amazon_US, WooCommerce) without an enum constraint. This invites hallucinated or misspelled channel names. Should declare valid channels as an enum: ["Shopify", "Amazon_US", "WooCommerce", ...].
Tool description does not explicitly state that this operation is destructive (generates purchase orders, modifies inventory state). Agents need to know: is this retryable? Can it cause duplicate POs? Idempotent or not?
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 33 | 2026-07-28+ | v2 |
No error handling guidance. The source code shows no try-catch, no validation, no error responses. If an invalid channel is passed, if API calls fail, or if a partial sync occurs, the agent has no recovery path.
'low_stock_alerts' response field is an array but no pagination guidance is provided. If there are hundreds of low-stock SKUs, returning all of them in a single response wastes tokens and risks context overflow. Should accept limit/offset parameters and return a total_count.
Tool description does not clarify dependencies or expected preconditions. Does the tool require channel credentials to be pre-configured? What if a channel is unreachable? Should agents call a setup or validate tool first?