Loading ToolBench
An MCP server deployed on Cloudflare Workers with OAuth 2.0 authentication support, demonstrating remote MCP server capabilities with a simple arithmetic tool
Single tool 'add' has a clear verb-noun name and basic schema with typed parameters. However, the description is extremely minimal (23 chars), barely meeting the 10-char floor but well below the 50-200 char LLM-optimized baseline. Parameter descriptions are present but terse (12-18 chars each). No output schema is documented, the response structure is inferred from code but not declared in the tool definition. No error handling guidance, no idempotency hints, no composition context. The tool is trivial (arithmetic), so some gaps are forgivable, but a production server would require richer metadata.
Adds two numbers together
Tool description is 23 characters, below LLM-optimized baseline (50-200 chars). Lacks context on when to use, what it returns, or any prerequisites.
Output schema is not documented. Response structure (content array with text type) is inferred from code but not declared in tool metadata. LLMs cannot plan downstream calls without knowing return fields.
Parameter descriptions are terse (12-18 chars). 'First number to add' and 'Second number to add' are adequate but do not specify format, range, or constraints. No mention of numeric bounds or precision.
No error handling guidance. Tool does not document what happens on invalid input (non-numeric, NaN, Infinity, overflow). LLM has no recovery path.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 47 | 2026-07-28+ | v2 |
No idempotency or composition hints. Tool does not declare whether it is safe to retry or how its output chains to other tools (though in this case, there are no other tools).