dsh-prompt-for-me
Run the following command in DeepSeek Harness:
dsh plugin install XXXXXQ-0206/dsh-prompt-for-me
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install XXXXXQ-0206/dsh-prompt-for-me in your terminal; the source code is available at https://github.com/XXXXXQ-0206/dsh-prompt-for-me . Restart dsh web after installation to activate the plugin.
About this plugin
Writing a good coding-agent prompt is a design task, not a casual chat. You need the model to understand the problem you are solving, what the project already contains, which file or module is the right next target, and which constraints must stay untouched. dsh-prompt-for-me moves that design work right next to the Harness composer, eliminating the repetitive cycle of manually gathering context and then drafting instructions.\n\nClick the button: if the composer holds a draft, it reads the project's bounded context (workspace, file tree, key manifests, recent git status and diff) and honestly optimizes your prompt, preserving your original intent without drifting into unrelated territory. If the composer is empty and a project is open, it designs a complete prompt for the next high-value engineering step. Results stream into the editor; a second click aborts and restores the original draft, while Ctrl+Z and Ctrl+Y let you walk through generated history. You always decide when to send.\n\nBuilt for engineers who drive day-to-day development through coding agents in DeepSeek Harness and want to turn a rough idea into an executable instruction in one step. It never auto-sends, never overrides your judgment, and simply gets the context and design work ready when you need it.
Use Cases
- Turning a rough draft into an executable instruction enriched with project context
- Designing a complete prompt for the next engineering step when the composer is empty
- Quickly converting a fuzzy requirement into a structured coding-agent instruction without repeated manual iteration
Best For
- Engineers driving day-to-day development through coding agents in DeepSeek Harness
- Developers who want to cut down on manual context gathering and prompt iteration
- Tech leads who need to turn fuzzy requirements into structured instructions quickly
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.