MCP server that wraps the XAIP Trust API — check scores, select servers, report results.
Four tools with clear names and reasonable descriptions, but schemas lack depth and parameter documentation is sparse. xaip_list_servers and xaip_check_trust have empty input schemas ({}), which violates schema completeness. xaip_select has a well-structured input schema with enums and descriptions. xaip_report has good parameter coverage but lacks output schema documentation. Descriptions are adequate (100-150 chars) but could be more prescriptive about when to use each tool. No error handling guidance visible in descriptions.
Check the trust score of an MCP server. Returns trust level, execution count, and risk flags.
List all MCP servers that have trust scores. Shows trust level, execution count, and verdict for each server.
Report the result of an MCP tool execution. Contributes to the trust score of the server. Each report is Ed25519-signed.
Select the most trustworthy MCP server for a task from a list of candidates. Automatically excludes unscored servers.
xaip_list_servers has empty input schema ({}). Tools with no input parameters should still document what they return and any prerequisites.
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract required fields without knowing response structure.
xaip_report description does not clarify that this is a WRITE operation with side effects (contributes to trust scores). Agents need to know which calls are idempotent.
No error handling guidance in any tool description. E.g., xaip_check_trust should explain what happens if a server slug is not found or unscored.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 56 | 2026-07-28+ | v2 |
xaip_select 'mode' parameter enum is documented but lacks guidance on when to use 'relative' vs 'strict'. Agents need decision criteria.