dsh-bromine: A Pure Node Implementation Without Network Dependencies
dsh-bromine is a pure Node.js plugin developed specifically for DeepSeek Harness (DSH) and runs without an internet connection. It is maintained by user uckkk and is suitable for deployment in intranet or offline environments. The installation command is `dsh plugin add github:uckkk/dsh-bromine`; before use, ensure the `@deepseek-ai/cordis` and `@deepseek-ai/dsh-tools` dependencies are installed. The plugin is released under the MIT license. It is recommended to read the source code before use to confirm its behavior and logic.
Read MoreIntroducing the DeepSeek Harness Plugin: dsh-phosphorus
dsh-phosphorus is a pure Node.js plugin for DeepSeek Harness (DSH), designed to provide local computational capability without network dependencies. The plugin runs entirely locally and does not call third-party APIs, making it well suited for building agents in offline or intranet environments. Installation requires the `dsh plugin add` command and depends on specific versions of `cordis` and `dsh-tools`. The plugin is open-sourced under the MIT License; it is recommended to review the source code logic before installation.
Read Moredsh-guangzhou-city: A Network-Independent Guangzhou City Information Query Plugin
The DeepSeek Harness plugin `dsh-guangzhou-city` is designed to address the pain point of relying on external APIs during DSH development. Built with pure Node.js, it can run locally without internet access and focuses on retrieving Guangzhou-related information. Developers can easily install it and invoke it directly in conversations, making it suitable for intranet or network-restricted environments. Before use, the source code’s security and license should be checked.
Read Moredsh-aluminium (Aluminium Element)
The DeepSeek Harness plugin `dsh-aluminium` is designed to give agents pure Node.js local computation capabilities, delivering network-independent functionality within the “everything is a plugin” philosophy. It is maintained by uckkk, with all runtime logic executed locally, making it suitable for workflows that require network isolation. Installation is performed by running commands in the terminal. Before use, ensure your environment meets the required dependency versions, including `@deepseek-ai/cordis` (^4.0.1) and `@deepseek-ai/dsh-tools` (>=0.1.0-rc.6). Because it contains core logic, it is recommended to review the source code and license before adoption. There are no concrete usage examples available yet; for more details, refer to its GitHub repository.
Read More