MCP server with MCP Apps integration
This server demonstrates significant quality gaps across naming, descriptions, schemas, and error handling. The two tools are minimally documented, lack comprehensive input schemas, and provide no guidance for error recovery. While the tools follow a basic verb_noun pattern, descriptions are superficial (under 100 chars), parameter types are incomplete, and output schemas are not documented. The server uses the mcp-use framework which abstracts some MCP mechanics, but source code visibility is severely limited, tool definitions cannot be directly verified from the provided code snippets. The framework appears to delegate actual tool registration to imported functions (e.g., registerProductSearch), but the actual schema definitions are not shown, forcing inference. Per the hard scoring rules, inferred tool definitions cap individual tools at 50; combined with missing/incomplete documentation across both tools, the overall score reflects poor definition quality.
Get detailed information about a specific fruit
Search for fruits and display the results in a visual widget
Tool definitions cannot be verified from provided source code; actual tool registration logic in mcp-use framework is opaque
Input schemas lack type definitions, length constraints, and pattern validation
Descriptions lack WHAT, WHEN, and dependency information required for LLM tool selection
Output schemas not documented; LLM cannot plan downstream tool calls or extract fields
No error handling guidance; tool does not indicate when fruit not found, invalid search, or other failure modes
Inferred effective spec: 2025-06-18+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 44 | 2025-06-18+ | v2 |
| 2026-03-09 | F | 41 | - | v1 |
Tool names use hyphens instead of underscores (search-tools, get-fruit-details)
Parameter descriptions missing or absent; 'query' and 'fruit' lack guidance on valid input