One authenticated MCP endpoint fronting every service in your homelab — built on Cloudflare Code Mode.
The server exposes a single tool 'dynamic_openapi_tools' that is dynamically generated from merged OpenAPI specs. The tool definition is inferred rather than explicitly registered in source code, only a reference to 'openApiMcpServer adapter' appears in src/lib/mcp-server.ts. No explicit tool registration, schema, or description is visible in the provided source. The tool name is generic and does not follow verb_noun convention (should be something like 'call_homelab_service' or 'execute_openapi_endpoint'). Without seeing the actual schema generation logic, parameter definitions, or output structure, the tool cannot be scored above the inference cap of 50. The description provided ('Tools dynamically generated...') is vague and does not explain WHAT the tool does, WHEN to use it, or what parameters it accepts.
Tools dynamically generated from merged OpenAPI specifications of configured homelab services. Each service contributes its endpoints as callable tools through the openApiMcpServer adapter.
Tool definition is inferred, not explicitly visible in source code. Only a reference to 'openApiMcpServer adapter' exists; actual tool registration, schema, and parameter definitions are not shown.
Tool name 'dynamic_openapi_tools' is generic and does not follow verb_noun convention. Should start with an action verb (e.g., 'call_', 'execute_', 'invoke_') to signal intent to LLMs.
Tool description is vague and does not answer: What does it do? When should the LLM call it? What parameters does it accept? What does it return? LLMs cannot determine when to select this tool.
No input schema visible for the tool. Cannot verify parameter types, descriptions, constraints, or whether required parameters are documented. Schema score must be 0.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 29 | 2026-07-28+ | v2 |
No output schema documented. LLMs cannot know what fields to expect from the response, making downstream tool chaining and data extraction error-prone.