DSH Hub
dsh-multitask cover

Senguk520/dsh-multitask

dsh-multitask

UIWeb UI40 GitHub stars· updated 2026-10-23

A Multitask "shock absorber" plugin & agent preset for DeepSeek Harness (DSH): the main chat only plans, briefs subagents and summarizes results, while file reads, commands, search and builds all run inside subagent contexts. Four worker types, an approval gate, zero DSH source changes.

Install

npx @deepseek-ai/dsh plugin --profile <profile> add dsh-multitask

Restart `dsh web` after install. Bundle APIs can change during the developer preview.

README badge

dsh-multitask DSH Hub badge
[![DSH Hub](https://dshhub.dev/badge/dsh-multitask.svg)](https://dshhub.dev/plugins/dsh-multitask)

Paste this into your README. The star count updates with every catalog sync.

From the README

Excerpt from Senguk520/dsh-multitask, cleaned of badges and images.

dsh-multitask

<p align="center"><b>The Multitask "shock absorber" mode for DeepSeek Harness — the main conversation never touches your codebase.</b></p>

<a href="README.md">English</a> · <a href="README.zh.md">中文</a> · <a href="docs/DESIGN.md">Design notes and measured evidence</a> · <a href="LICENSE">MIT License</a>

A work mode for DeepSeek Harness(DSH) in which the main conversation does exactly three things: understand what you asked for, turn it into task briefs, and report back what the Subagents concluded. Every piece of process data — file contents, command output, build logs, search results, dead ends — is digested inside a Subagent's context and never enters the main one. The main conversation therefore holds only high-value information.

This is a mechanism, not a prompt asking the model to behave. Once the mode is selected, the main conversation (the coordinator) genuinely has a narrower tool surface: the tools that would drag in process data — writing files, running commands, searching the web — are not in its visible set. It cannot call them, so it cannot pollute its own context. A Subagent (worker) is the opposite: it gets the full tool surface, does the work, and returns only its conclusion.

Zero runtime dependencies. No changes to DSH source.

At a glance:

CapabilityWhat it gives you
Narrowed coordinator tool surfacethe main conversation cannot write, run commands or search — those tools are genuinely absent, not merely discouraged
Process data never reaches the main conversationfile contents, command output, build logs and dead ends stay inside a worker's context
Four kinds of Subagenta full-surface worker, a fork that inherits completed turns, a programmatic PTC worker and a read-only worker, chosen by task shape
Dispatch to another modehand a brief to any other installed agent preset and receive only its conclusion
Two-level approval gatesafe commands run untouched; dangerous ones raise a card in the main conversation
Corrections take effect immediatelya follow-up message stops the worker's current turn first instead of queueing behind the wrong work
Mode isolationseveral modes can share one workspace; each narrows only its own sessions
Cross-session settingspick a model per worker kind and set the parallelism limit, both in Settings
Automatic workspace permission repairon Windows the one missing permission is added at session start, and only that one
Read-only metadata tool file_infobyte counts and file kinds without reading contents, so the coordinator can verify a claim itself
Zero runtime dependenciesno changes to DSH source

How it differs from other modes

Main conversation (coordinator)Subagent (worker)
Tool surfaceread-only checks + delegation + result retrieval + interactionfull (write files, run commands, build, search…)
Contextyour request + task briefs + conclusionsall the process data
Jobunderstand, translate, dispatch, verify, reportdo the work

What the coordinator keeps (a whitelist):

…

Related plugins