Loading ToolBench
MCP server with OAuth authentication and a simple arithmetic tool
Single tool 'add' has critical definition gaps. Input schema present with Zod types (a, b as numbers) but parameter descriptions are null, violates the requirement that every parameter needs a description. Tool description is missing entirely (no docstring visible in registration). Output schema not documented. No error handling guidance. Naming is acceptable (verb_noun pattern) but insufficient to overcome missing descriptions and documentation.
Tool 'add' has no description. LLMs cannot determine when or why to select this tool without a docstring explaining its purpose, when to use it, and what it returns.
All input parameters (a, b) have null descriptions. Parameter descriptions are mandatory, LLMs cannot infer that 'a' and 'b' are operands for addition without explicit text.
Output schema is not documented. The tool returns text content but the LLM has no formal specification of the response structure, forcing it to guess at field names and types.
No error handling or recovery guidance. If the tool fails (e.g., invalid input), the LLM receives no actionable error message or next steps.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | 2026-07-28+ | v2 |
Tool is trivial (bare arithmetic) and does not demonstrate real-world agent patterns. No pagination, no resource references, no composition with other tools.