Model Context Protocol server for Infoset ticket management
Server has 9 tools with complete Zod schemas and descriptions. Naming follows verb_noun convention (list_, get_, search_, create_, update_). However, descriptions are generic (avg ~80 chars, below 194-char baseline), lack WHEN-to-use guidance, and omit output schema documentation. Parameters are typed but lack constraints (enums, ranges, format specs). No error recovery guidance. Write operations (create_ticket, update_ticket) lack confirmation/dry-run patterns. Output is raw JSON dumps without field documentation.
Create a new Infoset ticket
Get contact information by contact ID
Get email thread content by email ID
Get SLA breach data for an Infoset ticket
Get single Infoset ticket detail by ID
Get activity logs for an Infoset ticket
List Infoset tickets with optional filters and pagination
Output schemas not documented. Tools return raw JSON.stringify() with no field descriptions. LLMs cannot plan downstream calls or extract required IDs (e.g., ticket_id for chaining).
Parameter descriptions lack constraints. 'status' accepts array of numbers but no enum or range documented. 'sortDir' accepts 'asc' or 'desc' but no enum constraint. LLMs may pass invalid values.
Write operations (create_ticket, update_ticket) lack confirmation or dry-run support. Agents can accidentally create/modify tickets without verification. No recovery guidance if operation fails.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | 2026-07-28+ | v2 |
Search Infoset tickets by keyword and optional filters
Update an existing Infoset ticket
Tool descriptions are generic and under 100 chars (baseline 194). Missing WHEN-to-use guidance. E.g., 'Get single Infoset ticket detail by ID' does not explain when to use get_ticket vs list_tickets or search_tickets.
Error handling returns raw API errors without recovery guidance. No categorization of retryable vs user-fixable errors. LLM cannot determine next action on failure.