dsh-image-gen
Run the following command in DeepSeek Harness:
dsh plugin install hoyyang/dsh-image-gen
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install hoyyang/dsh-image-gen in your DeepSeek Harness terminal to complete the installation; the source code is available at https://github.com/hoyyang/dsh-image-gen
About this plugin
DeepSeek Harness already handles conversation, reasoning, and task orchestration with ease, yet a simple request like "draw me a picture" still had to be offloaded to an external tool. dsh-image-gen closes that gap: the agent calls any image-generation gateway right inside the conversation, turning a one-liner into a finished image without switching windows or copy-pasting prompts.
The philosophy is three fields and you are done. Point it at any OpenAI-compatible image gateway, enter the model name and an API key, save, and it works immediately with zero code changes. Configuration can live in the visual settings page (effective on save), environment variables, profile config, or a local file—pick whichever fits your workflow. Triggering is flexible: use the /dsh-image-gen command for deterministic calls, say "generate a cyberpunk city poster" in chat and let the agent decide the timing, or invoke it as a tool from a script. Output size and quality are both parameterizable.
If you are already running agent workflows in Harness and have a self-hosted or third-party image gateway (one-api, new-api, official endpoints, etc.) and want image generation to be a natural, low-friction step in the flow rather than extra overhead, this plugin is built for you. Security is the default—keys never leave the machine, output directories are locked to 0700, and errors are auto-redacted. Costs stay predictable: one image per call and no automatic retries on failure, so generation is both effortless and under your control.
Use Cases
- Generate an image with a single line in a Harness conversation
- Invoke image generation on demand from a script or pipeline
- Route through a self-hosted one-api or new-api gateway for unified output
Best For
- Developers building agent workflows in DeepSeek Harness
- Teams running a self-hosted or third-party OpenAI-compatible image gateway
- Harness users who want code-free, save-and-go image generation setup
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.