dsh-i-have-adhd: DeepSeek Harness Always-On ADHD-Friendly Output System Prompt Plugin
`dsh-i-have-adhd` is a DeepSeek Harness client plugin designed to globally inject an ADHD-friendly SystemPrompt through a Host-only architecture, remaining persistently active without explicit loading on the model side. The plugin is based on an MIT-licensed rule set and supports configuring injection order and additional parameters. Developers can easily enable it via a YAML configuration file. It is suitable for scenarios that require consistently maintaining a specific output structure, such as action-first responses or numbered steps. Note that this is a community product and requires a Node.js 22+ environment.
Read Moredsh-i-have-adhd: ADHD-Friendly Output Shaping Plugin for DeepSeek Harness
`dsh-i-have-adhd` is a DeepSeek Harness plugin designed to reshape model output formatting, eliminate redundancy, and improve information scanning efficiency. By injecting system prompts, it enforces “action-first” structured responses. It provides the `adhd_on/off/status` tools and file markers, supporting persistent state and real-time toggling. The plugin’s rules have lower priority than DSH’s own rules; once enabled, model responses will directly include numbered steps, time estimates, and confirmation steps, making it suitable for efficiency-driven development.
Read MoreDeepSeek Harness Plugin: dsh-i-have-adhd
The DeepSeek Harness (DSH) plugin `dsh-i-have-adhd` is designed to address verbose agent output and unclear steps by reshaping the output style to suit rapid execution. Ported from `i-have-adhd`, it includes 10 core rules such as acting first, numbering multi-step tasks, providing specific time estimates, and suppressing digressions. Developers can install it via a command and enable the mode with `/i-have-adhd`, making agent responses more focused on concrete instructions and progress, suitable for scenarios that prioritize efficient execution.
Read More