A universal goal-tracking engine MCP server with roadmap management, goal tracking, and progress history capabilities.
Journey MCP exposes 3 tools with minimal documentation. Tool names follow verb_noun convention (get_roadmap, update_item_status, add_goal), which is positive. However, descriptions are extremely brief (10-30 chars), parameter descriptions are entirely absent, and output schemas are not documented. The schema for update_item_status and add_goal are visible in JSON Schema format with proper types and enums, but get_roadmap has an empty input schema. No error handling guidance, no recovery paths, and no indication of side effects. This server falls well below production baseline (194 chars avg description, 72 chars avg param annotation, 100% of A+ tools have param descriptions).
Add a new goal to a layer
Get the current Journey roadmap data
Update status of an item (done/pending)
All tool descriptions are under 20 characters and provide no context for LLM selection. 'Get the current Journey roadmap data' (34 chars) lacks WHEN to use it, WHAT it returns, or prerequisites.
No parameter descriptions exist for any tool. LLMs cannot infer that 'type' means layer vs milestone, or that 'status' must be 'pending' or 'done'. Parameter names alone are insufficient.
Output schemas are not documented. Callers do not know what fields get_roadmap returns, what structure update_item_status confirms, or what add_goal yields. This forces LLMs to guess downstream field names.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 41 | 2026-07-28+ | v2 |
No error handling or recovery guidance. If update_item_status fails (item not found, permission denied, write error), the LLM receives no actionable next step. Errors should guide retry, user input, or alternative tools.
Destructive operations (update_item_status, add_goal) lack confirmation or dry-run support. Agents can irreversibly modify roadmaps without safeguards. No idempotent hints or destructiveHint annotations present.