dsh-req-workbench
Run the following command in DeepSeek Harness:
dsh plugin install Songran241/dsh-req-workbench
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install Songran241/dsh-req-workbench in DeepSeek Harness to install this plugin; the full source repository is at https://github.com/Songran241/dsh-req-workbench .
About this plugin
In AI-assisted workflows, feature requirements often scatter across chat turns, pasted text snippets, and workspace documents, making them hard to track and revisit. dsh-req-workbench turns this fragmented output into a structured requirements workbench inside the DeepSeek Harness Web UI sidebar, where you can create, edit, delete, and check off requirements and subtasks so that conversation output instantly becomes a manageable, trackable backlog.
It supports three import sources: a one-click extract button beneath every assistant message parses that turn into a structured requirement list, and you can also paste Markdown text or read a workspace file with a preview-and-edit step before committing. Every subtask accepts a deadline, and the system surfaces persistent reminders in the sidebar and an overlay during the final 24 hours, polling every minute so nothing slips through.
Built for teams that use DeepSeek Harness to break down and follow up on requirements. Whether you are pulling to-dos straight from a conversation or batch-importing an existing document into deadline-bound subtasks, it turns vague backlogs into a clear, trackable workflow.
Use Cases
- Extract feature requirements from AI conversations and break them into subtasks
- Batch-import Markdown documents into a deadline-bound requirement list
- Check near-deadline subtasks in the sidebar and act before timeout
Best For
- Engineers using DeepSeek Harness to decompose and track requirements
- Product managers who need to crystallize task lists from AI conversations
- Dev teams that want a unified view of requirements scattered across chats and docs
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.