dsh-plugin-novelai
Run the following command in DeepSeek Harness:
dsh plugin install ysysz0343-web/dsh-plugin-novelai
Paste the following prompt into your AI chat to install this plugin:
Install it in DeepSeek Harness by running dsh plugin install ysysz0343-web/dsh-plugin-novelai (source: https://github.com/ysysz0343-web/dsh-plugin-novelai). It is pure JavaScript with no build step; restart dsh web after installation to activate.
About this plugin
NovelAI Diffusion V4.5 produces stunning anime-style images, but its prompt system is built on Danbooru tag conventions—assembling base_caption, character tags, negative prompts, and tuning parameters by hand is a real cognitive load for creators. dsh-plugin-novelai removes that friction: you describe the scene in plain natural language inside a chat, and the plugin's built-in prompt-design rules automatically break it down into structured fields (base_caption, characters, size_preset) before calling nai_generate_image to render the final PNG into your local workspace—no more opening a browser to fill in form fields one by one.
Two auxiliary tools round out the workflow. nai_tag_suggest queries official Danbooru tags by Chinese, English, or pinyin input, so the AI can reference verified tag spellings for characters, poses, camera angles, and environments instead of guessing from memory. nai_reverse_prompt takes a local image, sends it to any OpenAI-compatible multimodal model you configure, and reconstructs a reusable NAI-standard prompt from the visual content—handy for iterating variations in the same style.
The entire plugin is pure JavaScript with a built-in MessagePack binary-stream decoder, so there is no build step and it installs out of the box. It is well suited for AI-art enthusiasts working inside DeepSeek Harness who prefer a conversation-driven creative flow and want to spend their energy on visual imagination rather than tag engineering.
Use Cases
- Describe a scene in plain language and let the AI compose a structured NovelAI V4.5 prompt to render an image
- Query official Danbooru tags in Chinese or pinyin to normalize character and camera descriptions
- Reverse-prompt an existing image into an NAI-standard prompt and batch-iterate style variations
Best For
- Anime and AI-art creators who prefer a conversational workflow
- NovelAI artists who want to skip manual prompt assembly
- Developers integrating image generation into their DeepSeek Harness setups
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.