Agent collaboration · DeepSeek Harness plugins — Page 304

Look at video with your own eyes: a DeepSeek Harness plugin that samples frames with ffmpeg and returns them to the agent as real images, each labelled with its timestamp — no vision API, no API key, nothing leaves the machine.

Opt-in DeepSeek Harness skill bundle that turns course, lecture, tutorial, documentary, meeting, and talk videos into structured study notes.

AI video plugin for the dsh web GUI: multi-vendor video generation channels (OpenAI-compatible /v1/videos, Kling, MiniMax Hailuo, Volcengine ARK, custom generic async), FFmpeg video processing (probe / frame extraction / GIF / image compress), a lightweight storyboard studio (templates + shots + FFmpeg compose), Agent tools and a sidebar AI 视频 panel.

dsh-vidfetch — a DeepSeek Harness plugin to download videos.

Multi-provider AI video/image generation plugin for DeepSeek Harness (DSH). ComfyUI-style node canvas studio + multi-provider settings, with agent tools list_video_models / generate_video / poll_video_task. | DeepSeek Harness AI 视频/图片生成插件：多提供者、节点画布工作室、Agent 工具与 REST RPC。

Host-level vision bridge for text-only models: analyze_image tool (Ollama local / Xiaomi MiMo cloud / any OpenAI-compatible endpoint) returning structured evidence.

DeepSeek Harness 识图插件：为 Agent 注册 image_analyze 工具，把本地图片或 URL 发送给 OpenAI 兼容视觉模型（默认 OpenRouter google/gemini-3.7-flash）并返回文字分析

dsh-vision — a DeepSeek Harness plugin (tool template).

Give DeepSeek Harness eyes: a vision_analyze tool that sends an image (local path or URL) to an OpenAI-compatible vision model and returns a text description.

DSH Computer Use (vision + desktop control, auto-detects Windows native or WSL): see the Windows screen through the agent vision channel, drive mouse/keyboard with a Codex-style blue overlay (Esc to abort), plus a composer status pill.

Vision 台架：会话区调试/上位机点表（按段添加地址），共享任务与时间线，并安装「台架模式」Agent 预设。

Give text-only DeepSeek Harness agents vision — multi-backend (OpenAI Vision API + CDP bridge), cross-platform, cached, configurable.

DeepSeek Harness 多模态视觉桥：贴图自动转文字描述（llm/stream 代理）+ view_image/ocr_image 主动视觉工具 + 原生多模态路由自动跳过（rc.7 适配），让 text-only 的 DeepSeek 模型看见图片

Local-first vision for text-only DeepSeek Harness agents: route image understanding to a local OpenAI-compatible vision model, return structured JSON evidence (summary / OCR / layout / semantics / visual / uncertainty).

常驻视觉服务：直连视觉模型（默认 opencode-go/minimax-m3，回退 zai-coding-cn/glm-4.6v）。describe_image / subagent_vision 工具 + 粘贴图片自动转译（llm/stream 钩子）+ 输入框视觉状态小胶囊与详情页（活动日志：指令/思考过程/输出）。零子代理、零 agent 上下文开销，按会话记忆窗支持视觉追问与验收。

Vision for text-only DeepSeek Harness agents: delegate image reading to a one-shot subagent on a configurable vision route (MiniMax / Kimi / any OpenAI-compatible provider), keeping image bytes and the vision model's context out of the main session.

Point at a webpage, give your DSH agent feedback, and compare the result in place.

Visual plan mode for DeepSeek Harness: structured plan.json + plan.md, an editable React Flow canvas, comments, plan diff, versioned revisions, and reliable write-back to the agent.

DeepSeek Harness (dsh) bundle plugin: vision_analyze tool lets text-only LLM agents read images via SenseNova VLM, with Schemastery config and single-source credentials

Voice messages for DeepSeek Harness web UI: an Edge TTS listen button under assistant replies plus a voice_say tool the agent can use proactively

Voice for DeepSeek Harness — give text-only DeepSeek ears and a mouth: browser-native speech input (STT) + read-aloud (TTS), plus Whisper/TTS agent tools.

Voice alert plugin for DSH: plays voice, shows popup and system notification when a task completes or the agent asks the user

Voice input for DeepSeek Harness: a microphone seat inside the composer plus a Web settings page, transcribing through the browser engine or any OpenAI-compatible /audio/transcriptions endpoint.

Real-time duplex voice for DeepSeek Harness: Volcengine streaming ASR/TTS, agent reply narration, barge-in, wake word, live captions. | 实时双工语音插件：火山流式 ASR/TTS、回复朗读、打断、唤醒词、实时字幕。