Critical failures across all dimensions. The server exposes only 2 tools (nx_plan_start, nx_task_add) with minimal definitions visible in src/mcp/server.ts. No input schemas, parameter descriptions, or output schemas are present in the provided source. Tool descriptions are trivial (10-20 chars). The codebase imports from '@moreih29/nexus-core/mcp' but actual tool registration logic is not visible, definitions appear to be generated or delegated to an external package. Per HARD SCORING RULES: tools with no visible input schema score 0 on schema; descriptions under 20 chars score 0-20 on description. Without seeing explicit tool definitions, per-tool scores are capped at 50, but the absence of any schema or parameter documentation in the provided source forces them lower.
Start a new planning session
Add a task to the current plan
No input schemas visible for either tool. Per HARD SCORING RULES, schema score must be 0 when no schema is present in source.
Tool descriptions are trivial (10-20 characters: 'Start a new planning session', 'Add a task to the current plan'). Current descriptions lack parameter guidance, prerequisites, and return value documentation.
No parameter descriptions visible. LLMs cannot infer what 'plan' or 'task' parameters mean, what format they expect, or what constraints apply.
Tool definitions appear to be delegated to '@moreih29/nexus-core/mcp' (imported but not shown). Actual registration, schemas, and parameter details are not visible in src/mcp/server.ts. Per HARD SCORING RULES: if tool definitions are inferred rather than directly visible, cap per-tool scores at 50. Since no schemas are visible at all, scores are lower.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 30 | 2026-07-28+ | v2 |
No error handling guidance. Tools marked as WRITE risk but no documentation on failure modes, recovery steps, or idempotency. Per pattern:recovery-guide, errors must tell the LLM what to do next.