AI Agent Hub
Back to skills
Pet Detection and Reporting icon

Pet Detection and Reporting

Life Service Updated 2026.08.29

Paste the following prompt into your AI chat to install this skill:

Install @user_bb47e3e3/pet-detection-analysis following https://skillhub.cn/install/skillhub.md

About this skill

Problem Solved

In home pet monitoring, users often have only a camera clip or screenshot and need to know whether cats, dogs, or birds appear, how many there are, and whether the result can be traced. Reviewing video manually is time-consuming, and recognition results, counts, and reports are easily scattered across chat history, making archiving difficult.

How the Skill Works

The skill turns media input into structured pet detection results:
- Input: local file paths or web URLs; video can be mp4, avi, or mov, and images can be jpg, png, or jpeg, with a maximum file size of 10MB.
- Detection scope: recognizes cats, dogs, and birds in the target area and outputs category localization, quantity statistics, and presence detection.
- Execution: calls the scripts.pet_detection_analysis workflow, using parameters such as --input, --url, --media-type, and --confidence-threshold to control input and output.
- History queries: uses --list to read past detection reports from the cloud API and outputs them as a Markdown table, without relying on local memory or manual summaries.

Boundaries and Notes

Results are intended as a reference for home pet monitoring and do not replace safety, legal, or veterinary judgment. Network URLs are downloaded by the API side, so local preprocessing is usually unnecessary. Identity association is handled internally and should not be requested from or exposed to users. After analysis, the cloud service deletes original video data; retention behavior should still be understood in terms of the service's API and policies.

Use Cases

  • Upload a living-room mp4 clip to confirm whether a cat or dog appears and export a structured report.
  • Provide a yard jpg screenshot to count dogs and birds and get localization or presence conclusions.
  • When asked to review historical reports, fetch the cloud report list and render it as a Markdown table.
  • Pass a public video URL with a confidence threshold to filter low-confidence results and output JSON.

Best For

  • Pet owners who need remote confirmation of cats, dogs, or birds in home camera screenshots
  • Home monitoring maintainers who archive pet clips and trace historical detection reports
  • Engineers who automate camera-based detection workflows for multi-pet households
  • Indoor security testers who need pet counts and locations in target frames