{
  "id": 10130572,
  "title": "We Had Skills, Experts, Connectors, and Projects. Why Build Plugins?",
  "url": "https://urgent.news/2026/09/27/we-had-skills-experts-connectors-and-projects-why-build-plugins",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-27T03:19:47.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/techflag/we-had-skills-experts-connectors-and-projects-why-build-plugins-1ijo"
  },
  "original_language": "en",
  "account": "WorkDSH, an open-source AI workspace built on DeepSeek Harness, offers reusable skills, specialist configurations, external connectors, and projects. However, the developers also decided to create a plugin-based product. The reasoning behind this decision became clearer when they examined their Projects page. A project in WorkDSH is used to keep a particular task, its materials, and its history together. In contrast, a developer writes a Projects plugin to create and manage projects. While both share a name, they address different issues. A project organizes work, while a plugin supplies reusable operations. The developers found that a project already provides continuity for a team, and existing capabilities can assist with tasks. The decision to create a plugin arises when the agency requires consistency in formal quotes, such as confirming hotel prices before they appear on a quote. In such cases, a quoting plugin can provide a consistent record, editing screen, and approval behavior for multiple customer projects. It can also call a connector for fresh prices, suggest alternatives from a skill, and leave the final approval to an authorized person. However, if an agency already possesses a reliable booking and quoting system, it may be more efficient to connect that system instead of developing a new plugin. WorkDSH utilizes DeepSeek Harness's plugin runtime and published extension APIs, allowing it to add features without maintaining a modified upstream Harness checkout. The platform can accommodate both small tools and larger plugins with pages, services, and rules governing the data they manage. Before developing a new application plugin, the developers consider whether the user can complete the task using the existing project and capabilities, if there is a suitable system that can be connected via a connector, or if the missing operation occurs repeatedly across projects and requires the software itself to manage records and rules. Therefore, a new application plugin is justified only in the third case.",
  "summary": "An AI workspace can already do a great deal with reusable skills, specialist configurations, external connectors, and projects. We built all four into WorkDSH , an open-source workspace built on DeepSeek Harness. So why did we also make it a plugin-based product? The answer became clearer when we looked at our own Projects page. A user creates a project to keep a particular task, its materials,…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}