DeepSeek Harness Developer Tools & Plugins

DeepSeek Harness plugin that bridges Feishu/Lark bot chats to DSH sessions with streaming interactive-card replies.

A DeepSeek Harness plugin bundle that adds a WeChat bridge, web wallpaper controls, image-analysis workflow, and a unified settings page.

DeepSeek Harness plugin for running OmicVerse/OmicOS bioinformatics analyses through a persistent local Python kernel.

A DSH Web UI plugin for browsing, searching, and safely previewing files in the current session workspace.

Custom background wallpaper plugin for the DSH desktop UI.

A DeepSeek Harness vision extension that routes image tasks to configured GLM or Qwen vision models while retaining DeepSeek for text requests.

A publishable DeepSeek Harness web plugin that adds a “Model Plus” settings page and model metadata synchronization.

Semantic audio notifications for DeepSeek Harness agent events, with Web UI controls and browser or host playback.

Native Windows notification plugin for DeepSeek Harness with system toasts, sounds, and a taskbar tray badge.

HTTP/API debugging tools for DeepSeek Harness agents.

Read-only health diagnostics for installed DeepSeek Harness plugins, with a model tool, Web dashboard, and JSON reports.

A native DSH vision tool that describes local image files using mimo-v2.5-series multimodal models.

DeepSeek Harness plugin for safety-bounded DG-LAB Coyote e-stim control through agent tools and a web panel.

DeepSeek Harness Cordis bundle and CLI for configuring verified or custom LoongPort OpenAI-compatible providers.

Local-first structured work-state checkpoints and safe handoffs for DeepSeek Harness agents.

Evidence-backed, human-approved project memory for DeepSeek Harness.

A DSH Web client plugin that plays a synthesized chime when an assistant finishes replying.

A DeepSeek Harness CLI profile bundle that adds workspace switching and one-shot task execution.

Local personal bookkeeping plugin for DeepSeek Harness with expense and income tracking, budgets, statistics, and CSV/HTML exports.

A DeepSeek Harness realtime voice plugin that connects Qwen or OpenAI speech providers for ASR, voice-driven Harness turns, and streamed TTS playback.

A DeepSeek Harness change-review center for capturing file edits, reviewing diffs, applying changes, and rolling them back.

DeepSeek Harness web UI plugin that estimates remaining task time, token usage, and per-model cost in the session header.

Web UI microphone input plugin that converts speech into an editable DeepSeek Harness prompt draft.

A Web GUI desktop pet plugin for DeepSeek Harness featuring Yuhao, an interactive fox character.

A DeepSeek Harness plugin that automatically enables a simplified no-setup chat experience with curated defaults.

A DeepSeek Harness plugin that lets text-only models understand pasted or uploaded images using MiniMax VLM through mmx-cli.

A DSH web UI plugin for detecting and mutually switching installed theme/skin plugins.

DeepSeek Harness status-monitoring plugin with web UI, health APIs, SSE alerts, Prometheus metrics, and configurable thresholds.

Native DeepSeek Harness memory plugin combining local-first storage with XMemo cloud synchronization.

A DeepSeek Harness Web GUI plugin that displays a draggable floating card for model-account balances.

DSH Web client plugin that adds a Trajectory Reader tab for per-user-round summaries of assistant activity.

A DeepSeek Harness content-creation bundle for capture, card rendering, publishing, and human review.

A DeepSeek Harness web UI plugin that adds a floating HUD chat bar with transcript streaming, session controls, and Picture-in-Picture detach mode.

Unofficial ChatGPT OAuth provider for DeepSeek Harness that adds the openai-codex model route and stores credentials in native OS keychains.

A DSH web settings plugin that exports and incrementally restores installed third-party plugins, custom skills, and DSH configuration as JSON snapshots.

DeepSeek Harness plugin that admits image attachments in the Web GUI and converts them to text markers for a vision-skill workflow on text-only opencode-go upstreams.