MCP Remote Auth Demo - An OAuth 2.0 authorization server for MCP (Model Context Protocol) with Cloudflare Workers
Single tool 'add' has a basic schema with typed parameters (a, b as numbers) and descriptions, but lacks depth expected for production. Tool description is minimal (11 words, 33 chars), below the 10 - 1024 char guideline sweet spot (194 char baseline). No output schema documented. No error handling guidance. No indication of idempotency, side effects, or when to use this tool vs alternatives. The tool is trivial (arithmetic) but even simple tools need complete, LLM-optimized descriptions. Parameter descriptions are present but terse ('First number to add' is adequate but could explain constraints, valid ranges, or format expectations).
Adds two numbers together
Tool description is too short (33 chars vs 194 char baseline). 'Adds two numbers together' lacks context for LLM selection, prerequisites, or when to use this tool.
No output schema documented. LLMs cannot infer what fields the response contains or plan downstream tool calls. Response structure is implicit (text content only).
No error handling guidance. What happens if inputs are NaN, Infinity, or out of range? How should the LLM recover? No recovery hints provided.
Parameter descriptions lack constraint details. No mention of valid ranges, precision limits, or format expectations (e.g., 'integers only' vs 'floats allowed').
Tool name 'add' is generic and lacks verb_noun clarity. While 'add' is a verb, 'add_numbers' would be more explicit and self-documenting for LLM disambiguation.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 47 | 2026-07-28+ | v2 |