dsh-universe-api: Offline API Discovery Tool
`dsh-universe-api` is an offline API discovery plugin for DeepSeek Harness (DSH). It registers the `universe_api_search` tool, enabling deterministic search across 1,693 normalized public API records. The plugin supports offline retrieval, multilingual matching, and hard filtering (such as authentication, HTTPS, and CORS), and provides stable ranking. Users can filter APIs using natural language or parameters; however, this tool is only a discovery mechanism and does not perform network requests or invoke APIs, so users must verify specific services themselves.
Read Moredsh-visual-hierarchy: Visual Hierarchy Construction Reference
The community plugin `dsh-visual-hierarchy` for DeepSeek Harness (DSH), maintained by uckkk, is designed to provide an offline, network-independent solution for building visual hierarchy. Implemented purely in Node.js, it does not call external APIs. Users need to install it via the command line and enable it in `package.json`, after which the tools it provides can be used in DSH sessions. It is recommended to review the source code before installation to ensure security.
Read Moredsh-san-marino: San Marino
`dsh-san-marino` is a pure Node.js plugin for DeepSeek Harness (DSH), designed to provide local logic processing capabilities for agent development without network requests. It is suitable for use in restricted or offline environments and can be installed with the `dsh plugin add github:uckkk/dsh-san-marino` command. It depends on `@deepseek-ai/cordis` and `@deepseek-ai/dsh-tools` and should be run with the process's current permissions.
Read Moredsh-lanthanum: DSH Plugin for Lanthanum
`dsh-lanthanum` (lanthanum element) is a plugin for DeepSeek Harness (DSH) designed to provide a pure Node.js solution without requiring network access. The plugin runs entirely locally, does not depend on external network resources, and is especially suitable for restricted network environments. Before installing, it must meet the dependencies `@deepseek-ai/cordis` (^4.0.1) and `@deepseek-ai/dsh-tools` (>=0.1.0-rc.6). The installation command is `dsh plugin add github:uckkk/dsh-lanthanum`. The plugin is licensed under the MIT License, and it is recommended to review the source code before use.
Read Moredsh-http-status: HTTP Status Code Quick Reference
`dsh-http-status` is a local HTTP status code lookup plugin for DeepSeek Harness. It works without an internet connection and explains status codes such as 200 and 403 locally, preventing privacy leaks and reducing latency. After installation, it must be configured and enabled in the profile, making it well suited for offline scenarios or privacy-focused environments.
Read Moredsh-homework-help: Homework Help Plugin
`dsh-homework-help` is a homework tutoring plugin in the DeepSeek Harness (DSH) ecosystem, maintained by uckkk. It is implemented in pure Node.js with no network dependencies and can run in offline environments. The plugin accepts homework-related parameters and returns targeted tutoring suggestions. It is released under the MIT License and is suitable for use cases requiring structured homework tutoring locally or in restricted network environments.
Read Moredsh-guid: DeepSeek Harness GUID Generation Plugin
`dsh-guid` is a DeepSeek Harness (DSH) plugin maintained by developer uckkk, focused on generating globally unique identifiers (GUIDs) locally. The plugin is implemented in pure Node.js, does not depend on external network interfaces, and runs directly within the DSH process context. Users can install it using the `dsh plugin add` command and invoke its registered tools in a session. Since the plugin runs with the current process's permissions, please ensure the source code is secure. For specific interfaces, refer to `cordis.patch.yml` or the README in the source code.
Read Moredsh-chile: Chile
The dsh-chile plugin for DeepSeek Harness is a pure Node.js module maintained by uckkk, described as “Chile.” Its core features include a Node.js-based build, no runtime network dependencies, and suitability for offline environments. Developers can install it via a specified command, and it is recommended to review the source code before enabling it. Overall, it provides a lightweight localized module solution under the MIT license.
Read More