Preface¶

For requirements related to home surveillance, pet monitoring, or indoor security, the common approach is to integrate models like YOLO yourself and write the logic for video frame extraction and alerts. The model categories need to cover cats, dogs, and birds, and distinguish between single-image inference and continuous video stream detection, making integration costs significant. Here, we introduce the community Skill on SkillHub, “Pet Detection Skill”: it encapsulates the capability of “determining if cats, dogs, or birds appear within a target area” into an installable Agent Skill. It is designed for home pet monitoring scenarios and is marked as requiring no API Key (requires_api_key=false).

What is this¶

Pet Detection Skill is a life service (life-service) Agent Skill with the coordinate @user_bb47e3e3/pet-detection-analysis. The current public version is 1.0.7, the source is community, and it is maintained by the publisher user_bb47e3e3.

Official Description (Consistent in Chinese and English): Detects cats, dogs, and birds appearing in target areas; supports video stream and image detection; suitable for home pet monitoring scenarios.

On the SkillHub public page, this Skill has a rating of 100000 and 731 downloads (page data is subject to SkillHub and may change over time).

Core Functions¶

The following capabilities are derived from the SkillHub skill page and scraped metadata; interface names or configuration items not appearing in public materials are not extended.

1. Target Categories: Cats, Dogs, Birds.
2. Input Forms: Supports Image detection and Video Stream detection.
3. Scenario Positioning: Home pet monitoring; emphasizes whether the aforementioned animals appear within a “target area.”
4. Dependencies & Keys: Tagged with requires_api_key=false, meaning the capabilities described by the skill can be completed without relying on external API Keys (specific implementation is subject to the SKILL.md in the installation package).

Installation & Activation¶

SkillHub recommends two installation paths: providing the prompt to the current AI Agent for execution, or installing via command in an environment where the SkillHub CLI is already installed. Both methods point to the same coordinate and the same installation documentation.

Method 1: Install via Agent (Official Prompt)¶

The prompt provided on the SkillHub skill page is as follows; it can be sent directly to clients supporting Skills, such as Cursor, Claude Code, or Codex:

Please install `@user_bb47e3e3/pet-detection-analysis` according to https://skillhub.cn/install/skillhub.md.

The Agent will check for the CLI, select the correct skills directory, and complete the installation in accordance with SkillHub Installation & Priority Source Configuration.

Method 2: SkillHub CLI¶

First, confirm if skillhub exists on your local machine:

command -v skillhub && skillhub --version

If not installed, you can perform a full installation (CLI + Default Skill) according to the official documentation:

curl -fsSL https://skillhub-1388575217.cos.ap-guangzhou.myqcloud.com/install/install.sh | bash

When installing the skill, you must use --dir to point to the current Agent’s skills directory; if installed to the default ./skills/, some clients may not recognize it.

skillhub install user_bb47e3e3/pet-detection-analysis --dir <skills directory>

Common client skills paths (excerpted from official installation documentation):

Directory paths vary by client; please refer to your current environment. If the client does not automatically load the new skill after installation, restart or refresh the Skills list according to the product instructions.

Typical Usage¶

The public SkillHub page does not provide independent HTTP examples, request body fields, or local command-line demos; the reproducible “usage” is mainly reflected in the installation and activation process.

1. Complete the installation as described in the previous section, enabling the Agent to read the SKILL.md (general Skill format) within the Skill package.
2. Define the task boundaries in the conversation: for example, judging whether there are cats/dogs/birds in an indoor surveillance screenshot, or continuously monitoring a video stream for the appearance of the aforementioned animals within a target area.
3. The Agent invokes the capabilities encapsulated by the skill according to the skill documentation; specific parameters and output structures are subject to the SKILL.md after installation. This article does not fabricate unpublicized field names.

If you only require the capability boundary of “home scenario, three types of pets, image + stream,” you can directly reference the four types of information in the official introduction (categories, area, image, video stream) in your requirement description to avoid deviating from the skill’s design scope.

Use Cases & Notes¶

Who is it for

• Users already using Agents with Skills like Cursor, Claude Code, or Codex, who wish to quickly add pet detection capabilities rather than connecting models and pipelines from scratch.
• Requirements for “whether cats, dogs, or birds appear” in scenarios like home pet monitoring and indoor activity monitoring, accepting the maintenance rhythm of community-sourced (community) skills.

Limitations & Notes

• Public materials do not specify detection accuracy, frame rates, hardware requirements, or whether custom ROI coordinate formats are supported; verification should be performed on real footage (lighting, angles, occlusion) before deployment.
• Community Skill versions and download counts will be updated; prior to installation, verify on SkillHub whether v1.0.7 and the page description match the citations in this article.
• The skill description focuses on cats, dogs, and birds; other animals or “pet behavior analysis” (eating, fighting, etc.) are not within the verified introduction scope and should not be assumed to be supported.

Conclusion¶

Pet Detection Skill consolidates the detection of cats, dogs, and birds via images and video streams into an installable Agent Skill. Marked as requiring no API Key, it serves as a suitable starting point for home pet monitoring requirements. Installation and coordinates are subject to SkillHub.

• SkillHub Skill Page: https://www.skillhub.cn/skills/user_bb47e3e3/pet-detection-analysis
• Installation Instructions: https://skillhub.cn/install/skillhub.md
• Coordinate: @user_bb47e3e3/pet-detection-analysis