dsh-plugins
Run the following command in DeepSeek Harness:
dsh plugin install DIV7NE/dsh-plugins
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install DIV7NE/dsh-plugins in the DeepSeek Harness terminal to install this plugin; the source is available at https://github.com/DIV7NE/dsh-plugins .
About this plugin
The DeepSeek Harness web interface is already quite polished, but daily inference workflows still carry a few noticeable frictions: the next prompt is a pure hand-feel, browser automation demands a separate profile, and code blocks in chat can only be copy-pasted into a terminal by hand. dsh-plugins is a collection of self-contained web plugins built to close exactly those gaps. Each one is an independent package; upon install the CLI appends it to the current profile bundle stack automatically, with no manual config edits required.
Each plugin handles one job. dsh-suggest-next-prompt surfaces a model-generated next-prompt shortlist as the composer placeholder, with Tab to paste, arrow keys to cycle, and Esc to dismiss, keeping pace with your thinking. dsh-chrome-agent drives your already logged-in Chrome window through a lightweight companion extension, eliminating the need to maintain a second browser session. dsh-run-in-terminal adds a Run button beside every code block; clicking it opens an integrated terminal in the right sidebar, pastes the snippet, and hits Enter. Selecting terminal output and right-clicking appends it as conversation context, closing the loop from execution to understanding.
Built for developers and researchers who use DSH daily for model inference and want the shortest possible chain from prompt to runnable code to result read-back, without constant context switching between browser, terminal, and chat panels.
Use Cases
- Iterate prompts in DSH chat by instantly reviewing and switching model-suggested next prompts
- Drive an already-logged-in Chrome window for web automation or debugging without a separate profile
- Execute a chat code block with one click and feed the terminal output back into the conversation
Best For
- Developers who use DeepSeek Harness daily for model inference and want less context switching
- Researchers who need browser automation without maintaining a separate browser profile
- Full-stack engineers who want the shortest possible chain from prompt to runnable code to result read-back
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.