Loading ToolBench
MCP server for B2B trade marketplace operations including credit underwriting, logistics coordination, and trade order management between manufacturers and retailers
Critical failures across all dimensions. The single tool 'execute_skill_action' has a vague, generic description (58 chars) that does not explain WHAT the tool does, WHEN to use it, or what it returns. Parameters lack descriptions entirely, 'manufacturer_id', 'retailer_gstin', 'trade_value_inr' are named but unexplained. No output schema is documented anywhere in the codebase. The mcp_server.py file is a stub that prints JSON and exits; it does not implement the MCP protocol, handle tool calls, or process input. The actual business logic lives in client.py, which is never invoked by the server. This is not a functional MCP server, it is a skeleton with no protocol implementation.
Execute B2B trade marketplace credit underwriting and logistics skill action
Tool description is generic and uninformative (58 chars). Does not explain what the tool does, when to use it, or what it returns. LLMs cannot determine when to select this tool.
All three input parameters lack descriptions. 'manufacturer_id', 'retailer_gstin', 'trade_value_inr' are named but unexplained. LLMs cannot infer what values to pass or why each parameter matters.
No output schema documented. The tool returns structured data (trade_order_id, underwritten_credit_limit_inr, etc.) but the MCP server does not declare this schema. LLMs cannot plan downstream tool calls or extract required fields.
mcp_server.py does not implement MCP protocol. It prints a JSON status message and exits. It does not read stdin, parse tool call requests, invoke the business logic, or return structured responses. The server is non-functional.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 22 | 2026-07-28+ | v2 |
Tool name 'execute_skill_action' is vague and generic. Does not start with a clear action verb (get_, create_, update_, search_). 'execute_skill_action' could mean anything, no LLM can infer intent from the name alone.