In the DSH ecosystem, processing visual data often requires connecting to dedicated inference services. The following section introduces a plugin focused on flame recognition. By calling the XiaoBenYang API, it provides agents with flame detection capabilities.

Plugin Positioning

The plugin is maintained by xby-skill and belongs to the model inference category. Its core function is to detect flames in various general-purpose scenarios. It is best suited for perspectives from security cameras and traffic cameras.

Core Features

The plugin provides four core APIs, each handling a different input data format:

  1. set_xby_apikey: Sets the API key in the conversation. This setting is automatically persisted and remains effective after the process restarts.
  2. fire_detection: Detects flames. It requires an image file URL.
  3. fire_detection_for_data_base64: Detects flames. It requires the Base64 encoding of an image file.
  4. fire_detection_for_data_file: Detects flames. It requires a local file path to the image file.

Installation and Activation

Use the official installation command to add the plugin:

dsh plugin --profile web add github:xby_skill/xby-fire-detection

After installation, the plugin can be used in the conversation context.

Typical Usage

Before use, complete the following steps:

  1. Register on the official XiaoBenYang website and obtain an API key.
  2. Use the set_xby_apikey command in the conversation to set the key.
  3. Based on how the image is stored, call the corresponding detection function:
    - URL based: Call fire_detection and pass the image URL.
    - Base64 based: Call fire_detection_for_data_base64 and pass the Base64 string.
    - Local file: Call fire_detection_for_data_file and pass the file path.

Use Cases and Notes

  • Use cases: Scenarios that require real-time or offline flame detection, such as security monitoring and traffic intersection monitoring.
  • Note: The plugin depends on an external API, so you must obtain the key yourself. Because the plugin runs with the current dsh process permissions, it is recommended to check the source code and license (MIT) before installation.

By following the steps above, you can integrate flame detection directly into a DSH workflow. For more details, consult the plugin directory or the GitHub repository.