sk-doubao-seedream：在 Agent 里调用火山方舟 Seedream 文生图
sk-doubao-seedream（v1.0.1，@user_1060c188/sk-doubao-seedream）是 SkillHub 社区 design-media 类图片生成 Skill，封装火山方舟 Ark 上 Doubao Seedream 与 SeedEdit 模型。文章说明其文生图、图生图与多模型能力，给出 skillhub install、Cursor/Claude/Codex skills 目录、config.json 与 VOLCENGINE_API_KEY 配置，以及 generate.py 的 list、generate、edit 示例命令与适用注意。
Read Moredsh-ark-quota：火山方舟订阅额度侧边栏小组件
`dsh-ark-quota` 是一个专为 DeepSeek Harness 设计的火山方舟额度监控插件。它通过侧栏卡片直观展示订阅剩余额度，支持按模型路由自动切换账号，并提供“燃烧节奏”与耗尽预测。插件具备多账号管理、密钥本地校验及设置页热加载功能，解决了多模型切换下手动查询额度繁琐的问题，适合关注成本的开发者使用。
Read MoreDeepSeek Harness 插件：接入火山方舟 Agent Plan 与 Coding Plan
DeepSeek Harness（DSH）Ark 计划插件为 DSH 提供了火山方舟 Agent Plan 与 Coding Plan 的完整接入方案。它通过适配器声明 Provider，实现了模型接入、额度追踪及精细控制。 安装后需重启 profile，并在设置中配置 API Key。插件利用 `arkcli` 从控制面获取真实额度数据，在标题栏显示进度。此外，该插件支持模型管理（启用/停用），并内置了 Agent Plan 的 12 个模型 ID。已知限制包括不支持图片输入、无法调节思考强度及 Provider 删除困难，但整体解决了原生 DSH 对 Ark 套餐支持不足的问题。
Read MoreDeepSeek Harness Plugin: dsh-ark-quota (Volcano Ark Usage Monitoring)
`dsh-ark-quota` is a plugin for DeepSeek Harness (DSH), designed to enable real-time monitoring of Volcano Ark quotas from the web interface. By reading IAM credentials and calling the OpenAPI, it provides a right-side floating window (supporting multi-tier queries for Coding/Agent Plan), settings-page configuration, and local persistent storage. It supports automatic refresh, keeps AK/SK in local memory for security, and offers a model tool for conversational invocation. Installation requires manually adding the plugin package to `node_modules` and registering it in `cordis.patch.yml`; it takes effect after restarting DSH.
Read Moredsh-provider-veark: DeepSeek Harness Integration with Volcano Ark Coding Plan
`dsh-provider-veark` is the Volcano Ark plugin for DeepSeek Harness (DSH), designed to address the lack of native support for Coding Plan. The plugin supports multimodal understanding of text, images, and PDFs, and implements optimized image uploads and secure API key storage through the official API. Installation requires restarting DSH and manually allowing the `protobufjs` build script to run. After configuration, users can use the document citation feature via the web interface or the Composer button. The plugin runs with DSH process privileges and supports advanced YAML configuration, but it has limitations, including no support for video or audio and a 45 MiB limit for both single-file and cumulative sizes.
Read MoreDeepSeek Harness (DSH) Sidebar Plugin: Real-Time Display of Volcano Ark Agent Plan Usage
The DeepSeek Harness (DSH) plugin `dsh-ark-plan-usage` adds a visual card to the bottom of the Web UI sidebar, displaying real-time Agent Plan usage. The plugin retrieves data by calling the local `arkcli` and supports expand/collapse, color-coded warnings, and one-click re-login. With zero build, zero runtime dependencies, and a caching mechanism, it eliminates the need for developers to repeatedly switch to the terminal to check quotas.
Read MoreDeepSeek Harness Plugin: dsh-ark-balance
dsh-ark-balance is a widget-style plugin designed to address the pain point of checking the free resource package balance of the Volcano Ark Collaboration Reward Program in a Web interface. It reuses the official CLI data source, requires no additional login, and supports drag-and-drop layout, multi-dimensional filtering and sorting, and overlay metric display. The plugin features smart low-balance alerts, auto-refresh, and graceful degradation, ensuring guided onboarding when the CLI is not installed. It is a pure Node.js ESM module with zero dependencies and easy installation.
Read Moredsh-llm-volcengine: Integrating DSH with Volcano Ark Agent Plan and Coding Plan
dsh-llm-volcengine is an MIT-licensed DSH plugin maintained by Badakonpro, designed to simplify the integration of Volcengine Ark Agent Plan and Coding Plan. By registering two LLM provider routes, `volcengine-plan` and `volcengine-coding`, this plugin unifies the handling of protocol routing, compatibility toggles, credential parsing, and thinking intensity settings. It adopts a self-contained architecture and does not rely on any specific settings schema. Core features include: Agent Plan routing using the `openai-responses` protocol (with `openai-completions` for certain models), and...
Read More