Plugin

Cervello DeepSeek + trascrizione automatica delle immagini: allega immagini nella GUI e ognuna viene trascritta in testo tramite un qualsiasi VLM compatibile con OpenAI prima di raggiungere il DeepSeek solo testuale — un percorso rapido con chiave (predefinito qwen3.7-flash; DashScope/Zhipu/OpenRouter o qualsiasi endpoint compatibile con OpenAI) con la tua chiave, oppure Ollama locale rilevato automaticamente a configurazione zero.

A vision-language gateway provider route: pasted images are described by a configurable VL model (Qwen-VL by default) before the DeepSeek wire.

Supporto plugin multimodale per Qwen.

Bridge di visione nativo del provider LLM: le immagini incollate in chat vengono descritte da un modello di visione (Qwen3-VL via pi-ai/llama.cpp) e la descrizione testuale viene passata a DeepSeek solo testo per la risposta — ammissione delle immagini, instradamento e compattazione avvengono tutti tramite meccanismi nativi dell'harness, con una cache LRU delle descrizioni e retry sui 503

Universal image generation for DeepSeek Harness: auto-discovers image models from any OpenAI-compatible endpoint (SenseNova, StepFun, Agnes, Qwen, Flux, SD, Imagen and more), with agent tools and REST API.

Native vision capability extension, using either Zhipu (free) or Qwen-VL (local).

Renders pasted/uploaded images inline in the DSH Web chat and gives text-only models vision: the model-invokable visual_review tool calls any OpenAI-compatible multimodal API first, falling back to a local Qwen3-VL worker.

Free vision bridge for text-only models: image understanding, OCR, UI and debug analysis via free-tier providers (Qwen3-VL-Flash, Doubao, DeepSeek-OCR) with a settings GUI.

让 DeepSeek Harness 里任何纯文本模型都能读图。识图是按需调用的 tool —— 图片不进主模型上下文，不看就不花钱；附 23 处缺陷的评测集，换模型可自测。

DeepSeek Harness out-of-tree plugin: give a text-only coding model eyes by routing images to a Qwen-VL (DashScope) route through ctx.llm and returning text.

Dispatch self-contained tasks from DSH to 11 external agent CLIs (Codex, Claude Code, TraeCode, OpenCode, Gemini, Cursor, Kimi, Qwen, Copilot, WorkBuddy, Grok): a host-only bundle registering the conductor_dispatch tool plus an auto-triggered conductor skill.

Agent-initiated voice calls: `offer_call` rings the human (接听/拒接/稍后再说); accepted calls synthesize and play locally via CrispASR + Qwen3-TTS (9 speakers, 2 Chinese dialects), rejected calls return the decision to the agent.

On-demand vision for text-only DeepSeek models: upload images, and the model calls a view_image tool backed by any OpenAI-compatible vision endpoint (Qwen/DashScope by default).

Plugin per DeepSeek Harness: interruttore della modalità di pensiero per richiesta (enable_thinking) per i modelli della famiglia Qwen3 su endpoint compatibili con OpenAI

Bundle di plugin per DeepSeek Harness: strumenti qwen_vision (comprensione delle immagini con Qwen-VL) e qwen_generate (testo-immagine con Qwen-Image) per modelli solo testo

Model-agnostic visual-inspection pipeline for text-only agents: page-by-page HTML screenshots plus a ready-made vision-subagent briefing contract (gewu_prep), then source-code truth verification of every finding (gewu_locate); validated on mimo-v2.5 & qwen3.7-plus.

明眸 VisionBridge - 自研视觉桥：瞎子模型收图时自动调用视觉模型识别，把识别文字喂回主模型，无感、可配置、升级不丢

Voice AI girlfriend for the Web UI: FunASR mic input, Qwen3-TTS spoken replies, companion animation window, and two-way QQ chat (text/voice/image push) via NapCat.