Unofficial, local-first Comci timetable change intelligence CLI and API server for Korean school schedules
Server exposes 10 tools with basic schemas and descriptions. All tools have names starting with action verbs (search_, get_, list_) and descriptions present. However, descriptions are minimal (10-50 chars), parameter descriptions lack depth, and output schemas are not documented. No error handling guidance, no tool annotations, and no composition patterns evident. Tools are read-only but lack idempotent hints. Schema quality is moderate, all params have types but lack format constraints, ranges, or validation guidance. This is typical community-grade work: functional but not production-ready.
Get schedule changes within a date range
Get class schedule for a specific day
Get class schedule for a specific week
List teachers for a school on a specific date
Health check endpoint for liveness probe
Health check endpoint for readiness probe
Get JSON schema for a domain model
Descriptions are too brief (10-50 chars). LLMs cannot determine when to select tools or what they return. Baseline for A+ tools is 50-200 chars with WHAT/WHEN/WHY context.
Output schemas are not documented. LLMs cannot plan downstream calls or extract required fields. Every tool must declare what it returns.
No error handling guidance. Tools do not explain what to do on failure (retry, ask user, abort). Agents cannot recover from errors.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | 2026-07-28+ | v2 |
Search for schools by name
Get teacher schedule for a specific day
Get teacher schedule for a specific week
No tool annotations (readOnlyHint, idempotentHint). All tools are read-only but lack explicit hints. Agents cannot reason about safety.
Parameter descriptions lack format/range constraints. E.g., 'target_date' says 'YYYY-MM-DD format' but no validation guidance. 'grade' and 'class_no' lack min/max bounds.