I Have ADHD Skill · Productivity

I Have ADHD Skill: Make Agent Output Easier to Follow

Put the next doable action where the reader can see it.

Ayoub Ghriss's explicit-invocation Skill makes agent replies easier to act on: start with the next action, use bounded steps, keep state visible, and show progress. It is an output-formatting mode, not diagnosis, treatment, or medical advice.

View Ayoub Ghriss’s source

ADHD-friendly output
Clear next actions
User-invoked mode

About This Skill

I Have ADHD Skill makes agent replies easier to act on. Invoke it to put the first action up front, turn work into bounded steps, carry state across turns, suppress tangents, and end with one doable move.

Built on proven foundations

Trusted Source, Professionally Refined

The original expertise stays intact. Focused refinements make it easier to start, follow the guidance, and know when the work is finished.

Original foundation

Ayoub Ghriss · commit 4c76175

The original Skill knowledge, package structure, attribution and original license terms are preserved. Follow the source link to see where this edition began.

View the original project

Antigravity optimization

4 files, one clearer workflow

A source-faithful edition with the improvements explained below. The download includes the original attribution and everything needed to follow the Skill’s guidance.

More useful from the first prompt

What We Improved

Keep the depth you came for. Get clearer starting points, practical guidance and an explicit finish line.

Recognize when to use the mode

Choose this explicit mode when the task needs easier-to-follow output without changing its requirements.

Keep the task intact

Make replies easier to follow while keeping technical, safety, and user-requested requirements in charge.

Make handoffs actionable

For cross-turn work, name the current step, first artifact or path, uncertainty, and one next action.

Name the first thing to open

End a handoff with the first artifact or path to open so the next step is obvious.

Start with the Next Action

1Invoke `/i-have-adhd` when you want the session’s output shaped for easier action; the mode stays explicit until stopped.
2Put the first doable action on line one and number only the bounded steps the task actually needs.
3Restate current state each turn, suppress side issues, and give a concrete time estimate when work remains.
4For a handoff, name the current step, first artifact or path, uncertainty, and one next action before sending.

Example Command

Use /i-have-adhd. Give me the next action first, then three numbered steps, and end with one thing to do in under two minutes.

Core Capabilities

Next-action-first output

Put a small, concrete action before background context so the reader can start immediately.

Bounded numbered work

Use the fewest numbered steps that still work and avoid hiding multiple actions inside one step.

State carried across turns

Restate the current step and keep the selected mode active until the documented stop phrase.

Visible progress and handoff

Show completed work, concrete time estimates, and the first artifact or path for continuation.

Use Cases

Coding-agent execution

Put the command, path, or first edit before explanation while keeping technical literals exact.

Multi-step plans

Turn a long task into a short numbered path with one bounded action per step.

Cross-turn continuation

Carry current state and one next action through a later turn or agent handoff.

Focused explanations

Finish the first issue before offering a separate tangent or optional follow-up.

SKILL.md

# I Have ADHD

This focused edition derives from Ayoub Ghriss's MIT-licensed `i-have-adhd` Skill at commit 4c76175b17421d61babfc35c10ab07604ef578d0. It is an explicit-invocation output-shaping mode, not diagnosis, treatment, or medical advice.

The original ten rules remain the authority: lead with the next action, number real multi-step work, end with one concrete next action, suppress tangents, restate state, give concrete time estimates, make completed work visible, use a matter-of-fact error tone, cap visible lists, and avoid preambles or empty closers. The focused instruction diff adds clearer scope protection, a conditional handoff check, and a first-artifact pre-send check. Platform metadata remains explicit and optional hooks are not bundled here.

HumanizerA source-preserving writing pass for reducing formulaic prose patterns.Stop SlopA red-pen editing workflow for clearer AI prose without dropping source facts.CavemanA terse communication mode that protects technical literals and clarity.

Frequently Asked Questions

What is I Have ADHD Skill?

It is Ayoub Ghriss’s explicit-invocation Skill for ADHD-friendly output: obvious next actions, bounded numbered steps, state carried across turns, fewer tangents, concrete time estimates, and visible progress.

Is it a diagnosis, treatment, or medical tool?

No. The source explicitly frames it as output shaping. It does not diagnose ADHD, provide treatment, or replace medical advice.

How do I start or stop the mode?

Install the Skill from ayghri/i-have-adhd using its documented Skills path, then invoke `/i-have-adhd` when you want the explicit mode. Say “stop adhd mode” or “normal mode” to turn it off; host-specific controls remain the canonical repository’s responsibility.

How does it help with a coding task?

It puts the command, path, or first edit first, keeps multi-step work short and numbered, and ends with one concrete action while preserving technical literals.

Does it rewrite technical or safety requirements?

No. The mode shapes presentation while preserving technical, safety, and user-requested requirements. If clarity and a requirement conflict, the requirement remains authoritative.

Discussion

0 comments

Trigger Phrases

Use these phrases to activate this skill in your AI coding assistant:

/i-have-adhdMake this easier to follow.Lead with the next action and number the steps.Stop adhd mode.