MCP Gateway (beta): authn/z + tenancy + policy + transforms
Scoring was not performed
7 of 9 tools lack descriptions entirely or have empty descriptions. LLMs cannot determine when or why to select these tools.
Duplicate tool registration: 'whoami' appears twice in the tool list with conflicting metadata (different files/descriptions). This indicates test fixture pollution or malformed tool definitions.
6 of 9 tools have no visible input schema at all (empty {} or missing). Without schemas, LLMs cannot validate parameters or understand what arguments are valid.
No output schemas documented for any tool. LLMs need to know what fields to expect in responses to plan downstream operations and extract relevant data.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 33 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 21 | - | v1 |
Tool names are not action-oriented. 'echo_request', 'echo_headers', 'echo_query' sound like debugging utilities rather than production-grade tools. Names should indicate user intent (e.g., 'send_http_request', 'call_api').
The server is clearly a test/integration test fixture (all tools reside in crates/adapter/tests/ and crates/adapter/src/bin/). Production MCP servers should not expose test tools. These should be separated into a dedicated test server.