Loading ToolBench
MCP server with OAuth 2.0 provider for remote authorization
Single tool 'add' has a clear, actionable name and proper schema with typed parameters. However, the tool description is minimal (25 chars), falling short of the 50-200 char baseline for LLM-optimized descriptions. Parameter descriptions exist but are generic. Output schema is not documented. No error handling guidance. The tool is simple enough to work, but lacks the robustness and completeness expected for production use.
Adds two numbers together
Tool description is too brief (25 characters). LLM-optimized descriptions should be 50-200 characters, answering WHAT it does, WHEN to use it, and WHAT it returns. Current: 'Adds two numbers together' lacks context about return format and use cases.
Output schema not documented. LLMs need to know what fields to expect in the response to plan downstream calls. Current code returns {content: [{type: 'text', text: String}]} but this structure is not visible in the tool definition.
No error handling or recovery guidance. If invalid inputs are passed (non-numbers), the tool provides no actionable error message. Should validate inputs and return clear error classification (retryable vs user-fixable).
Parameter descriptions are minimal. 'First number to add' and 'Second number to add' lack format/range guidance. Should specify: 'A number (integer or decimal, any magnitude)' to clarify what inputs are valid.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 52 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 31 | - | v1 |