dsh-imagegen
Run the following command in DeepSeek Harness:
dsh plugin install little-traincar/dsh-imagegen
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install little-traincar/dsh-imagegen in your DeepSeek Harness terminal; the full source lives at https://github.com/little-traincar/dsh-imagegen .
About this plugin
When drafting copy, building a deck, or iterating on a product description inside a DSH Agent session, the need for a quick illustration, promo poster, or stylized graphic often forces you to leave the conversation, open another tab, type a prompt, wait for a render, then paste the result back. dsh-imagegen registers a generate_image tool directly on the agent so that Doubao Seedream 5.0 Pro or Qwen-image-3 becomes a capability the agent can invoke mid-conversation.
Describe what you need in plain language and the agent calls the API, embeds the image inline in the chat bubble, and saves it to a local directory automatically. If you already have a reference image, paste it into the session and add a short restyle instruction; the agent passes the image through the image parameter so the model re-renders based on that reference. Count, aspect ratio, timeout, and base URL are all configurable, and the gateway can point to any OpenAI-compatible /images/generations endpoint.
Ideal for solo creators, design-support workflows, and marketing teams that already live inside DSH Agent. All you need is an Ark (Doubao) or DashScope (Qwen) API key; paste it into the settings page, hit save, and the image capability is live with no extra server, no separate frontend, and no pipeline to maintain.
Use Cases
- Describe a poster or illustration in chat and let the agent invoke Seedream or Qwen to render it
- Paste a reference image and ask the agent to restyle, swap the background, or re-draw
- Batch-produce marketing assets with images embedded in-session and saved to a local folder
Best For
- Solo creators already working inside DSH Agent for copywriting and content
- Small teams that repeatedly need promo posters or design-assist graphics
- Developers holding a Doubao Ark or Qwen DashScope API key
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
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.