AI Agent Hub
Back to plugins
dsh-plugin-task-coordinator preview

dsh-plugin-task-coordinator

Workflow Updated 2026.09.12

Run the following command in DeepSeek Harness:

dsh plugin install Kayungko/dsh-plugin-task-coordinator

Paste the following prompt into your AI chat to install this plugin:

In DeepSeek Harness, install by searching task-coordinator in the in-app plugin marketplace, or clone from https://github.com/Kayungko/dsh-plugin-task-coordinator and run the install.ps1 deploy script, then restart DSH Desktop to activate the plugin.

About this plugin

When you push multiple sub-tasks in parallel inside DeepSeek Harness, the hard part is rarely model quality; it is orchestration: how to decompose work, who owns what, how to collect results, and how to track progress. dsh-plugin-task-coordinator wraps that glue into a structured toolset so you describe intent in plain language while the plugin handles decomposition, approval, batch spawning and result collection. It ships eleven tools covering discovery, deep-reading, message delivery, cancellation and more, with interactive approval cards and a live topology view that keep decision points in your hands. Built for developers who regularly run multi-session workflows in DSH and need a supervisor mode to coordinate sub-agents, especially when splitting a large development effort into parallel sub-tasks and integrating the outcomes.

Screenshots

Use Cases

  • Split a large development effort into parallel sub-sessions and integrate results centrally
  • Pass instructions across multiple AI sessions, track status, and adjust priorities on the fly
  • Confirm a decomposition plan via an approval card before batch-spawning to avoid blind parallelism

Best For

  • Developers who regularly run multi-session workflows in DeepSeek Harness
  • Engineering teams that need a supervisor-worker pattern to coordinate AI sub-agents
  • Power users who want a structured pipeline for task decomposition, approval, and result collection