Loading ToolBench
Prebuilt CLI, development server, and build pipeline for mcp-use
Critical gaps across all dimensions. Two tools with minimal descriptions (10-15 chars each, well below 20-char floor). Parameter descriptions are null/missing entirely, no type information visible for parameters despite schema presence. No output schemas documented. No error handling guidance. Tools appear to be test fixtures rather than production-grade definitions. Schema structure exists but lacks the descriptive richness required for LLM tool selection.
Add two numbers
Search products
All parameter descriptions are null. Parameters 'a', 'b', 'query' have no descriptions, violating the requirement that every parameter must explain what it controls. LLMs cannot infer parameter meaning from names alone.
Tool descriptions are under 20 characters ('Add two numbers' = 15 chars, 'Search products' = 14 chars). Descriptions lack context for LLM selection: when to use, what it returns, prerequisites.
No output schemas documented. LLMs cannot plan downstream tool calls or extract data without knowing what fields to expect. Both tools lack return type documentation.
No error handling guidance. Tools provide no recovery hints, error classification, or actionable error messages. An LLM receiving a failure has no guidance on retry, user correction, or next steps.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 40 | 2025-06-18+ | v2 |
Parameter 'query' in search-products is marked optional but no default is specified. Unclear what happens when omitted, does it return all products, error, or empty results?