Simple and easy-to-use AI assistant with knowledge base, model API, sharing, web search, and agents. Includes MCP server implementations for testing with STDIO and SSE transports.
Server has 4 tools with explicit schemas and descriptions. All tools are READ_ONLY and well-intentioned for testing. However, descriptions are minimal (10-80 chars, well below the 194-char baseline), parameter descriptions lack depth, and output schemas are not formally documented. Tools follow verb_noun naming but lack context about when to use each. No error handling guidance, no recovery paths, and no actionable error messages. Schemas use Zod validation but descriptions don't explain constraints (e.g., maxEntries 1-50 limit not mentioned in description text). This is a test server, which explains the simplicity, but production-grade tools would require richer documentation.
Echo input text and attach basic metadata.
Return the current server runtime status for smoke testing.
Inspect files under the current workspace for MCP integration testing.
Sum a numeric array for tool invocation testing.
Descriptions are too brief (10-80 chars vs. 194-char baseline). 'Return the current server runtime status for smoke testing' lacks context about when to call this vs. other tools, what fields are returned, or how to interpret the response.
Parameter descriptions lack constraint details. 'maxEntries' has a 1-50 range enforced in schema but the description text does not mention this limit. LLMs cannot read JSON Schema constraints, they rely on description text.
No output schema documentation. Tools return JSON text but the structure (fields, types, required vs. optional) is not formally declared. LLMs cannot plan downstream calls without knowing what fields to expect.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 51 | <=2025-11-25 | v2 |
No error handling or recovery guidance. If 'inspect-project' receives an invalid path, the error response does not guide the LLM on what to do next (e.g., 'Try listing the workspace root first').
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). All tools are READ_ONLY but this is not declared in the tool definition, only in the metadata. Modern MCP spec expects tool annotations.