Cat Litter Box Waste Characteristic Recognition

Problem It Solves

Smart litter boxes and multi-cat households generate many videos, but owners mainly care whether stool is soft, bloody, or hard, and whether urine clumps look unusually small or large. Reviewing video manually is slow and lacks a consistent record. This skill turns litter-box footage into standardized observations: stool form, color, clump size, excretion trends, and health-risk notes. It is positioned as characterization and risk flagging, not disease diagnosis.

How It Works

• Input: supports a local video file or a video URL; requires mp4 / avi / mov, up to 10MB.
• Core analysis: a built-in script processes the input, identifies stool form (diarrhea-like, bloody, or hard), color features, and clump size, then generates a structured report, monitoring results, suggestions, and a report link.
• History reports: when the user asks for past reports, report lists, or health-risk notes, the skill fetches the latest cloud data via --list and renders a Markdown table; it does not summarize from local memory or session records.
• Boundaries: identity parameters are handled internally and are not asked for or exposed; results are for health reference only and do not provide diagnosis or treatment advice; use the provided scripts instead of generating temporary replacement scripts.