← Back to home

Author

@niushuanan

DeepSeek Harness plugins published by @niushuanan on GitHub.

View on GitHub13 plugins

dsh-chat-migration

4 ★

Native DeepSeek chat migration and workspace-free chat mode for DeepSeek Harness

Memory

xiaozhuang-dsh

1 ★

Plugin-enhanced DeepSeek Harness distribution with Computer Use, Teamwork, model usage, and multi-worktree development.

Auth

dsh-pure-chat

1 ★

Start a chat immediately without a workspace, work mode, or execution permissions while keeping image and text-file uploads.

Web UI

dsh-whale-girl

1 ★

Add a native cross-page companion whose presence, shortcuts, and feedback follow the current DSH session state.

Fun

dsh-multi-window

0 ★

Open multiple independent conversations side by side, each with isolated navigation, drafts, and runtime state.

Web UI

dsh-skill-manager

0 ★

Browse installed Skills, their files, and content in Settings, then adaptively import from a file, folder, ZIP, or GitHub with AI.

Skills

dsh-adaptive-update

0 ★

Check upstream manually or every six hours, use a narrowly scoped agent for compatibility work, and switch atomically with rollback.

Tools

dsh-selection-memory

0 ★

Quote, discuss, or remember selected conversation text, then maintain durable context in separate editable user and AI memories.

Memory

dsh-model-usage

0 ★

Inspect model quotas, periods, and refresh state by provider in Settings, with data loaded only when the user opens the page.

Auth

dsh-teamwork

0 ★

Run collaborating agents and external experts concurrently under one coordinating agent, then bring every result back into the current task.

Workspace

dsh-parallel-worktree

0 ★

Move parallel tasks into isolated Git worktrees, inspect conflicts, and merge the results safely into the current branch.

Coding

dsh-token-overview

0 ★

See tokens, cache usage, calls, active periods, and estimated cost across AI clients on the whole computer.

Auth

dsh-image-vision

0 ★

Let vision-capable models read native attachments while giving text-only models an image tool that supports follow-up questions.

Vision