Introduction¶
The core design philosophy of DeepSeek Harness (DSH) is “everything is a plugin.” When developing agents for specific domains, it is often necessary to inject domain knowledge or specific tool capabilities into the model’s workflow. In the photography domain, obtaining professional composition advice is a common need. The uckkk/dsh-compose plugin was created to address this issue. It provides a lightweight solution that can return photography composition suggestions based on parameters.
Plugin Scope¶
This is a DSH plugin for model inference, maintained by user uckkk. Its core function is to return photography composition plans according to specified parameters. The plugin is implemented purely in Node.js and does not depend on network access, ensuring it works normally in offline or restricted network environments.
Installation and Dependencies¶
Before installing, please make sure the current environment meets the following dependency requirements:
@deepseek-ai/cordis(^4.0.1)@deepseek-ai/dsh-tools(>=0.1.0-rc.6)
Use the official installation command to add it:
dsh plugin add github:uckkk/dsh-compose
Usage¶
After installation, the plugin automatically registers tools. Directly invoke these tools in a session to obtain photography composition plans based on the specified parameters.
Notes¶
This plugin follows the MIT license. Because the plugin runs with the privileges of the current DSH process, it is recommended to review the source code before installation to ensure security and compliance.