DeepSeek Harness (DSH) adopts a plugin-based architecture, with various features extended through plugins. During local development or debugging, quickly obtaining available AI model information is a common need. The dsh-ai-models plugin provides a localized model query capability, allowing you to retrieve model entries and their details without an internet connection.

Core Features

The plugin mainly provides the following two core capabilities:

  1. List models: Use the list_aimodel tool to get the list of currently available AI models.
  2. Query details: Use the get_aimodel tool to retrieve detailed information for a specific model by ID or name.

Installation and Prerequisites

Before using the plugin, make sure your environment meets its peerDependencies requirements:

  • @deepseek-ai/cordis (^4.0.1)
  • @deepseek-ai/dsh-tools (>=0.1.0-rc.6)

The installation command is as follows:

dsh plugin add github:uckkk/dsh-ai-models

Usage

After installation, the plugin registers two tools in the session: list_aimodel and get_aimodel. Developers or agents can call these tools directly in conversations or scripts.

Technical Implementation and Notes

  • Implementation: The plugin is implemented in pure Node.js and does not rely on network requests, making it suitable for offline or intranet environments.
  • License: The plugin is open-sourced under the MIT License.
  • Security note: The plugin runs with the permissions of the current DSH process. It is recommended to review the source code and license information before installing.

The steps above complete the configuration of dsh-ai-models. If you need to access more model resources or participate in maintenance, you can visit its GitHub repository.

GitHub repository