AI work intelligence server that captures patterns, decisions, and bugs from AI coding sessions. Provides MCP tools for session analysis, grounding, and context graph building. Supports local LLM inference via Apple FoundationModels or MLX, with BYOK (Bring Your Own Key) analysis via LiteLLM.
Single tool 'nora_help' has minimal definition quality. Description is 115 chars (within acceptable range) but lacks actionable detail on WHEN to use it, WHAT it returns, or HOW to interpret results. No input schema visible in source code, tool appears to take no parameters, but this is inferred rather than explicitly declared. No output schema documented. Tool name 'nora_help' is generic and does not follow verb_noun pattern (should be 'list_tools' or 'get_available_tools'). No error handling guidance. The tool is read-only (low risk) but definition quality is well below production baseline.
List the MCP tools active on your tier (15 Lite / 60 Pro). Run in chat to see available tools.
No input schema visible. Tool definition inferred rather than explicitly registered with JSON Schema.
Tool name 'nora_help' is generic and does not follow verb_noun convention. Should be 'list_tools' or 'get_available_tools' to clearly signal action.
No output schema documented. LLM cannot plan downstream actions or extract structured data from response.
Description lacks WHEN to use, WHAT it returns, and HOW to interpret tier limits (15 Lite / 60 Pro). Incomplete docstring forces LLM to guess.
No error handling guidance. If tool fails or tier limit is exceeded, LLM has no recovery path.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | 2026-07-28+ | v2 |