What the project is and the problem it solves
The i-have-adhd project is an open-source skill designed for coding assistants, specifically to change how they present information when helping developers. Its core purpose is to prevent the assistant from burying the answer in lengthy explanations, pleasantries, or tangential details. Instead, it instructs the assistant to lead with the immediate next action, provide numbered steps for multi-step tasks, and end with one concrete next step. This approach aims to make interactions more direct and actionable, reducing the need for users to scroll through introductory text to find the useful part of the response.
Why it is gaining traction now
The project has attracted significant attention, as indicated by its high number of stars on the platform where it is hosted. This interest appears to stem from a growing recognition among developers and users of AI coding assistants that the default conversational style of many large language models can be inefficient for technical tasks. The model often begins with phrases like “Great question! Let me think about this,” offers background context, and ends with open-ended offers like “Hope this helps!” or “Let me know if you want to dig deeper.” For users seeking quick, precise coding assistance—especially those who may find such verbosity distracting or overwhelming—this pattern creates friction. The i-have-adhd skill offers a simple, rule-based way to override this behavior, making the assistant’s output more aligned with the needs of focused technical work. Its framing as “ADHD-friendly” (while noting no diagnosis is needed) has resonated broadly as a description of output that prioritizes clarity and action.
How it works at a high level
The project functions as a skill or plugin that modifies the behavior of a coding assistant. Users install it by following instructions provided in the project’s documentation, typically involving copying a configuration or command into their assistant’s interface. Once active, the skill applies a set of ten specific rules to the assistant’s responses. These rules dictate the structure and content of the output: leading with action, numbering steps, ending with a concrete next step, suppressing tangents, restating the current state, providing specific time estimates, making progress visible, stating errors factually, limiting lists to five items, and eliminating preambles, recaps, and closers. The skill does not alter the underlying AI model but instead guides how it formats and delivers its generated advice, effectively acting as a post-processing filter or behavioral constraint on the assistant’s output style.
Who it is for and real use cases
The skill is intended for anyone using a coding assistant who finds the typical verbose, conversational style unhelpful for getting straight to the point in software development tasks. This includes professional developers, students, and hobbyists working on code who value efficiency and clarity. Real-world use cases, as illustrated in the project’s documentation, involve common coding assistance scenarios like debugging, refactoring, or implementing features. For example, instead of receiving a response that starts with appreciation for the question, outlines multiple considerations, and ends with an open invitation to continue, the user gets a direct instruction: “Edit src/auth.ts:42 to update the token validation,” followed by numbered steps to open the file, replace a function, run a specific test, and a clear next step based on the test outcome. This format is particularly useful in iterative coding workflows where users need to quickly understand what to do next without parsing through explanatory fluff.
Honest caveats
The project is presented as an open-source initiative under the MIT license, meaning it is free to use, modify, and distribute, but comes with no warranties or guarantees of support. It appears to be primarily driven by individual contributors, as evidenced by the credits acknowledging inspiration from a book on adult ADHD while clarifying that the adaptation is for LLM behavior, not human organization. There is no mention in the provided material of formal maintenance schedules, institutional backing, funding sources, or long-term roadmap. Users should evaluate the skill’s compatibility with their specific coding assistant setup and be aware that its effectiveness depends on the assistant’s ability to follow the prescribed rules consistently. As with any community-driven tool, ongoing updates and issue resolution rely on contributor activity, which may vary over time.
Source: https://github.com/ayghri/i-have-adhd
