stdio bridge to the hosted VineVerse MCP server — the Bible as a knowledge graph
The server exposes only 1 tool (get_passage) with a minimal definition. The tool has a valid name starting with 'get_' verb, a brief description, and a single string parameter with a description. However, the parameter lacks type information in the visible schema, the description is very short (62 chars, below the optimal 10-1024 range), and there is no documented output schema. The implementation is a transparent proxy to a remote server, so tool definitions are not explicitly registered in the index.js file, they come from the remote endpoint. The definition quality is further hampered by the absence of error handling guidance, no pagination support documented, and no output schema documentation visible in the source.
Retrieve a Bible passage by OSIS identifier
Tool definitions are not explicitly visible in source code; they are inferred from the remote VineVerse server.
No output schema is documented. Users cannot see what fields get_passage returns, what format OSIS identifiers are expected to produce, or how to chain this tool's output into downstream operations.
Tool description is 62 characters ('Retrieve a Bible passage by OSIS identifier'). While not under 20 chars, it is terse and does not answer WHEN to use this tool, WHAT preprocessing may be required, or any error recovery guidance. Rubric baseline for tool descriptions is 194 chars (p10=34, p90=392).
Parameter 'osis' has a description ('OSIS (Open Scripture Information Standard) identifier for the passage, e.g. 'John.3.16'') that includes an example value. The rubric explicitly discourages example values in descriptions as LLMs reuse them literally. Replace the example with a regex pattern or format constraint.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 39 | <=2025-11-25 | v2 |
No error handling guidance. What happens if an invalid OSIS identifier is passed? What if the passage does not exist? The tool provides no recovery guidance to LLMs, violating the error classification pattern.