dsh-task-board
Run the following command in DeepSeek Harness:
dsh plugin install scwlkq/dsh-task-board
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install scwlkq/dsh-task-board in your DeepSeek Harness terminal to install this plugin, source: https://github.com/scwlkq/dsh-task-board
About this plugin
When multiple agent task rounds execute inside DeepSeek Harness, progress often lives only in chat transcripts, leaving no persistent board to track status, approval decisions, or execution history.
dsh-task-board bundles the Host service, Session bridge, Typert RPC contribution, and browser UI into a single installable package. On the board you can create durable task cards with titles, descriptions, acceptance criteria, workspaces, Agent Presets, and image attachments; edit, reorder, reopen, and delete cards; start and stop execution through ordinary DSH Sessions; review finished work, approve it, reject with feedback, send follow-up instructions, or retry a failed round; and inspect per-round activity alongside the full Session history, refreshing the browser projection from the authoritative Host snapshot.
It suits developers and teams who run agent workloads via the DeepSeek Harness Web profile and want a structured, auditable task workflow instead of reconstructing context from conversation threads.
Use Cases
- Manage status and approval decisions across multiple agent task rounds on a structured board
- Review agent output, reject with feedback, or retry a failed round
- Trace per-round Session history and activity logs without scrolling through chat transcripts
Best For
- Developers running agent workloads inside DeepSeek Harness
- Engineering teams that need to audit and track agent task progress
- Engineers who prefer a structured workflow over reconstructing context from chat threads
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.