prompt-polish
Run the following command in DeepSeek Harness:
dsh plugin install JOJO666888888/prompt-polish
Paste the following prompt into your AI chat to install this plugin:
In DeepSeek Harness, run dsh plugin install JOJO666888888/prompt-polish to install the plugin (source: https://github.com/JOJO666888888/prompt-polish), then restart dsh and hard-refresh your browser.
About this plugin
In dsh, a prompt draft is often just a fragment. prompt-polish solves exactly that: before you hit Enter, a dedicated prompt-engineering agent completes, disambiguates, expands, and structures the draft into a well-formed instruction with all key elements in place.
Core capabilities include one-click polishing (auto-filling goals, scope, constraints, output format, and acceptance criteria), multi-round rewriting with arbitrary version rollback, direct write-to-input or one-step send, and support for any OpenAI-compatible API including local runtimes such as Ollama, vLLM, and LM Studio. The plugin auto-mounts at harness startup with no build step and no per-session setup.
Ideal for developers who compose prompts frequently in dsh web mode and want to move past unstable model output caused by under-specified instructions. It reuses your existing default model routing in dsh, so no extra configuration is required.
Use Cases
- Polish a rough draft into a structured prompt before sending
- Iterate through multiple rewrites to converge on the best version
- Connect to local Ollama or vLLM without relying on cloud APIs
Best For
- dsh users who compose prompts frequently
- Developers using local inference runtimes
- Tech teams seeking more consistent prompt output
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.