Loading ToolBench
MCP server with OAuth 2.0 authentication provider for remote access, built on Cloudflare Workers
Single tool 'add' has critical definition gaps. Tool name lacks action verb (should be 'compute_sum' or 'add_numbers'). No tool description present. Input parameters 'a' and 'b' have null descriptions, violating the requirement that every parameter must have a non-empty description. Schema is present (Zod z.number()) but descriptions are missing entirely. Output schema not documented. This server fails multiple hard scoring rules: parameters lack descriptions (description score capped at 0), and the tool itself has no description (tool description score capped at 0).
Tool 'add' has no description. LLMs cannot determine when or why to select this tool without explicit guidance on its purpose and use case.
Input parameters 'a' and 'b' have null descriptions. Every parameter must have a non-empty description explaining what it controls. LLMs cannot infer parameter meaning from names alone.
Tool name 'add' does not start with an action verb. Should be 'compute_sum', 'add_numbers', or similar. Verb-first naming helps LLMs parse intent from the name alone.
Output schema is not documented. LLMs need to know what fields to expect in the response to plan downstream operations and extract data correctly.
No error handling guidance. If the tool fails (e.g., invalid input), there is no recovery path or actionable error message for the LLM.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | 2026-07-28+ | v2 |