Offline route planning service with terrain analysis, risk evaluation, pace estimation, and route selection capabilities
Server has 5 tools with basic descriptions and partial schemas. Tool names follow verb_noun pattern (nav.route, nav.risk_eval, nav.pace_estimator, nav.select, nav.export), which is good. However, descriptions are generic and lack actionable context for LLM selection. Input schemas are visible but incomplete: parameters lack detailed type constraints, enums, and validation guidance. Output schemas are not documented in the visible code. Error handling and recovery guidance are absent. The server is functional but falls short of production-grade quality expected for agent tooling.
Export selected route and analysis results to file formats
Estimate travel time and pace for routes based on terrain, mode of travel, and load
Evaluate risk factors (slope, exposure, hydrology) for generated routes
Generate route candidates between start and end coordinates using pathfinding algorithm with terrain analysis
Select optimal route from candidates based on risk evaluation, pace estimates, and constraints
Output schemas not documented. LLMs cannot plan downstream calls or extract required fields without knowing what nav.route, nav.risk_eval, nav.select, and nav.export return. Code shows route_to_dict() serialization but no formal schema definition visible.
Parameter descriptions lack actionable constraints. 'route_ids' is described as 'List of route IDs to evaluate; if empty, evaluates all routes' but does not specify format, length limits, or valid ID patterns. 'mode' in nav.pace_estimator lacks enum values (foot, bike, vehicle?).
Tool descriptions are generic and do not explain WHEN to use each tool or dependencies between them. nav.select says 'Select optimal route' but does not state that nav.route and nav.risk_eval must be called first, or what constraints are mutually exclusive.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | 2026-07-28+ | v2 |
No error handling or recovery guidance visible. If nav.route fails (invalid coordinates, no path found), the response does not tell the LLM what to do next or what to try instead.
nav.export accepts optional 'basename' but does not document file formats, output location, or what happens if export fails. LLMs cannot reason about side effects or retry strategy.