TabNexus-DSH
Run the following command in DeepSeek Harness:
dsh plugin install KaichenCurry/TabNexus-DSH
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install KaichenCurry/TabNexus-DSH in your terminal, then visit the source at https://github.com/KaichenCurry/TabNexus-DSH , restart the DSH Web service and refresh the page to start using the plugin.
About this plugin
When doing research or project work in DeepSeek Harness, tasks and web references tend to scatter across browser tabs and sticky notes, with no lightweight panel to tie them together. TabNexus-DSH fills that gap: it is a pure-UI, local-first DSH plugin that registers no Agent tools, requires no API key, and is ready to use the moment you click the top-right capsule.
Its core capabilities focus on four things: creating tasks with a clear goal; organizing web pages into renameable and moveable categories; maintaining a title, URL, note, and a three-state status (pending, in progress, done) for each page; and viewing overall progress through a simple three-segment flow view. All data lives exclusively in the browser local storage of the DSH Web Client, with no cloud upload, automatic tracking-parameter stripping, and duplicate prevention built in.
It is well suited for users who collect sources, manage references, or build a personal knowledge base inside DSH: no backend deployment, no Chrome extension, no model calls required, just open and go, with data persisted across sessions.
Screenshots
Use Cases
- Archive scattered web references into categories and track reading progress during literature review
- Record clear goals across multiple workstreams inside DSH instead of sticky notes
- Flip page statuses on the fly and get an at-a-glance view of overall progress
Best For
- Researchers who collect and organize web sources inside DSH
- Personal project managers wanting a lightweight task board without a backend
- Privacy-conscious users who prefer local-first tools with zero API key requirements
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.