What Is the Primary Purpose of the i-have-adhd Skill?

The i-have-adhd skill is a productivity-focused extension for coding assistants that reshapes every response to be ADHD-friendly, making output instantly actionable by leading with concrete next steps, numbering tasks, and providing specific time estimates.

The ayghri/i-have-adhd repository provides a coding assistant skill designed specifically to bridge the gap between knowledge and execution for users with ADHD. Unlike standard assistant responses that may bury action items in verbose explanations, this skill enforces a strict output format that prioritizes clarity, immediacy, and progress visibility. According to the source code in skills/i-have-adhd/SKILL.md, the implementation defines concrete rules that govern how the assistant structures every interaction once activated.

Core Objective: From Knowing to Doing

The primary purpose centers on eliminating friction between understanding a task and executing it. In skills/i-have-adhd/SKILL.md (lines 15-20), the documentation establishes that the skill remains active for the entire session until explicitly disabled, ensuring consistent formatting without requiring repeated activation commands.

The skill achieves this by filtering out distracting tangents and imposing a rigid structure that transforms vague suggestions into physical, bounded actions. This approach directly addresses executive function challenges by externalizing task sequencing and time management.

The Five Structural Rules of ADHD-Friendly Output

Lead with the Next Concrete Action

Every response must begin with the immediate physical action the user should take. As defined in skills/i-have-adhd/SKILL.md (lines 33-41), the first line should tell the user exactly what to do before any explanatory context appears. For example, the response starts with "Run npm install jsonwebtoken" rather than discussing the package's features.

Number Multi-Step Tasks as Single Actions

Complex workflows are decomposed into discrete, bounded steps. The rules in skills/i-have-adhd/SKILL.md (lines 42-55) mandate that each step represents one specific action, presented as a clean numbered list rather than paragraphs of instructions. This prevents the cognitive overload that occurs when users must parse dense text to identify their next move.

Provide Specific Time Estimates

Vague duration phrases like "a while" or "soon" are prohibited. Instead, skills/i-have-adhd/SKILL.md (lines 82-86) requires every step to include concrete time estimates in minutes or hours. This allows users to budget their attention and energy accurately, addressing time-blindness common in ADHD.

Restate State Each Turn

To prevent context loss common in ADHD workflows, the assistant must repeat the current step and progress status at each interaction. According to skills/i-have-adhd/SKILL.md (lines 73-78), this restatement ensures users never lose track of where they are in a multi-step process, even if they become distracted between messages.

Make Wins Visible

Completed actions receive explicit highlighting to provide immediate positive feedback. The specification in skills/i-have-adhd/SKILL.md (lines 89-95) emphasizes visual distinction for finished tasks, reinforcing progress and motivation through tangible evidence of advancement.

Activating and Deactivating the Skill

Users invoke the skill through simple command syntax defined in .opencode/command/i-have-adhd.md. Once activated, the behavior persists session-wide until explicitly terminated.

To activate:

/i-have-adhd

Typical response format after activation:

Run `npm install jsonwebtoken`, then edit `src/auth.ts:42`.

1. Open `src/auth.ts`
2. Replace the `verifyToken` function (lines 42-58) with the new snippet
3. Run `npm test -- auth.spec.ts`

≈ 5 minutes total. Next: run the tests and paste any failing line.

To deactivate and return to standard assistant behavior:

stop adhd mode

As noted in skills/i-have-adhd/SKILL.md (lines 15-20), accepted termination phrases include "stop adhd mode" or "normal mode".

Summary

• The i-have-adhd skill restructures coding assistant output to be action-oriented rather than explanation-heavy, targeting the gap between knowing and doing
• Five core rules govern the format: concrete action leading, numbered single steps, specific time estimates, state restatement, and visible wins
• Activation occurs via /i-have-adhd as defined in .opencode/command/i-have-adhd.md, persisting until "stop adhd mode" is issued
• Implementation resides primarily in skills/i-have-adhd/SKILL.md with strict line-specific behavioral definitions (lines 33-41, 42-55, 73-78, 82-86, and 89-95)

Frequently Asked Questions

What is the primary purpose of the i-have-adhd skill?

The skill transforms coding assistant responses into ADHD-friendly formats that prioritize immediate actionability over lengthy explanation. According to the ayghri/i-have-adhd source code, it helps users move directly from understanding to execution without cognitive overload by enforcing strict formatting rules that externalize task management.

How does the i-have-adhd skill differ from standard assistant modes?

Unlike default modes that provide contextual preamble before instructions, this skill mandates that the very first line contain the concrete next action. It also requires numbered lists for all multi-step tasks, specific time estimates for every action as defined in skills/i-have-adhd/SKILL.md (lines 82-86), and explicit progress restatement at each turn to prevent context loss.

How do you activate and deactivate the i-have-adhd skill?

Users activate the skill by issuing the /i-have-adhd command as defined in .opencode/command/i-have-adhd.md. The session remains in ADHD mode until explicitly turned off using phrases like "stop adhd mode" or "normal mode" as documented in skills/i-have-adhd/SKILL.md (lines 15-20).

Where are the formatting rules for the i-have-adhd skill defined?

The complete rule set—including requirements for leading with actions (lines 33-41), numbering steps (lines 42-55), providing time estimates (lines 82-86), restating state (lines 73-78), and highlighting completed work (lines 89-95)—is defined in skills/i-have-adhd/SKILL.md. Each behavioral constraint includes specific line references that govern the assistant's output format.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}