How to Activate the 'i-have-adhd' Skill in Your Coding Assistant Session: Complete Guide
To activate the 'i-have-adhd' skill, install the plugin for your coding assistant (Claude Code, Codex, Gemini CLI, etc.), then invoke it with /i-have-adhd or enable always-on mode with a flag file.
The 'i-have-adhd' skill from ayghri/i-have-adhd is a productivity-enhancing plugin that reshapes AI assistant responses for ADHD-friendly interaction. Once activated, it enforces 10 core rules—action-first formatting, bounded steps, concrete time estimates, and no preamble clutter—defined in [skills/i-have-adhd/SKILL.md](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md). This guide covers installation and activation across all supported runtimes.
How the 'i-have-adhd' Skill Architecture Works
The skill follows a consistent pattern across platforms, as documented in [AGENTS.md](https://github.com/ayghri/i-have-adhd/blob/main/AGENTS.md):
- Core rules: Single source of truth in
SKILL.md(10 ADHD-friendly response rules) - Runtime adapters: Platform-specific manifests in
.claude-plugin/,.codex-plugin/,opencode.json, etc. - Activation modes: Explicit invocation per session OR always-on via flag file
The always-on mechanism relies on a SessionStart hook (Claude Code) or equivalent auto-loading snippet (other agents) that injects rules when a flag file exists—~/.claude/.i-have-adhd-always for Claude Code, or analogous paths per runtime.
Activating the 'i-have-adhd' Skill by Runtime
Claude Code
Claude Code uses the plugin marketplace and slash-command activation.
# One-time installation
claude plugin marketplace add ayghri/i-have-adhd
claude plugin install i-have-adhd@i-have-adhd
# Activate for current session
/i-have-adhd
Enable always-on mode so the skill loads automatically:
touch ~/.claude/.i-have-adhd-always
Codex
Codex follows a similar plugin pattern with $ prefix activation.
# Install the skill
codex plugin marketplace add ayghri/i-have-adhd --ref main
codex plugin add i-have-adhd@i-have-adhd
# Activate for current session
$i-have-adhd
Gemini CLI
Two installation paths exist: command route (manual) or extension route (always-on).
Command route (manual activation per session):
# Download the command file
curl -fsSL https://raw.githubusercontent.com/ayghri/i-have-adhd/main/skills/i-have-adhd/agents/gemini.toml \
-o ~/.gemini/commands/i-have-adhd.toml
# Activate
/i-have-adhd
Extension route (automatic loading via GEMINI.md):
gemini extensions install https://github.com/ayghri/i-have-adhd
Pi
Pi supports explicit on/off toggling.
# Install
pi install https://github.com/ayghri/i-have-adhd
# Activate variants
/i-have-adhd # toggle current session
/i-have-adhd on # explicitly enable
/i-have-adhd off # explicitly disable
Oh My Pi (OMP)
omp plugin marketplace add ayghri/i-have-adhd
omp plugin install --scope user i-have-adhd@i-have-adhd
# Activate
/i-have-adhd
Qwen Code
qwen extensions install ayghri/i-have-adhd
# Activate
/i-have-adhd
Zed
Zed imports skills directly via URL to SKILL.md.
# Import (paste URL into Zed's skill import)
https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md
# Activate
/i-have-adhd
Generic Harness (Cursor, Amp, etc.)
For assistants without dedicated plugin systems, use the npx skills CLI.
# Workspace-only installation
npx skills add ayghri/i-have-adhd
# Global installation
npx skills add ayghri/i-have-adhd -g
# Activate
/i-have-adhd
Manual Always-On Setup for Any Assistant
If your coding assistant supports persistent rules files but lacks a formal plugin system, paste the core rules directly. Add this block to your agent's always-on rules file (e.g., ~/.cursor/rules/always.md, ~/.config/opencode/AGENTS.md):
## Output style
The reader has ADHD. Shape every response so it can be acted on:
1. Lead with the answer or next action: command, path, or snippet first.
2. Number multi-step work; one bounded action per step.
3. End with one next action doable in under two minutes.
4. Finish the current issue before raising a new one.
5. Restate progress each turn ("step 3 of 5 done").
6. Give time estimates in concrete units, never "a bit".
7. After a change, show what now works.
8. Errors: state location, cause, and fix. No drama.
9. Cap lists at 5 items.
10. No preamble, no recaps, no closers.
These 10 rules mirror the definition in [SKILL.md](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md) and activate the same behavioral changes without plugin infrastructure.
Key Reference Files in the Repository
| File | Purpose | Location |
|---|---|---|
SKILL.md |
10 ADHD-friendly response rules—the core rule set all runtimes implement | skills/i-have-adhd/SKILL.md |
INSTALL.md |
Platform-specific commands for install, verify, and update | Repository root |
AGENTS.md |
Runtime-to-entry-point mapping (Claude, Codex, Gemini, Pi, OMP, Qwen, Zed) | Repository root |
README.md |
High-level overview and quick-start pointer | Repository root |
Summary
- Install once: Use marketplace commands,
curl,npm, or direct URL import per your assistant. - Activate per session: Type
/i-have-adhd(or$i-have-adhdin Codex). - Enable always-on: Create a flag file like
~/.claude/.i-have-adhd-alwaysor use the extension auto-load path. - Fallback option: Paste the 10 rules manually into persistent rules files for unsupported assistants.
The skill activates immediately upon invocation, restructuring all subsequent responses to prioritize actionability, clarity, and ADHD-friendly formatting.
Frequently Asked Questions
What does the 'i-have-adhd' skill actually change in assistant responses?
The skill enforces 10 concrete formatting rules from SKILL.md: leading with actionable answers, numbering steps, capping lists at 5 items, providing concrete time estimates, eliminating preamble/recap chatter, and restating progress each turn. Responses become scannable, bounded, and immediately useful rather than verbose and open-ended.
Can I use the 'i-have-adhd' skill with Cursor or other assistants not officially listed?
Yes. Any assistant supporting persistent system prompts or rules files can use the skill. Either install via npx skills add ayghri/i-have-adhd for the generic harness, or manually paste the 10 rules into your editor's rules file (e.g., ~/.cursor/rules/always.md). The [INSTALL.md](https://github.com/ayghri/i-have-adhd/blob/main/INSTALL.md) file documents fallback methods.
Why does Claude Code use a flag file for always-on mode instead of a setting?
The SessionStart hook in Claude Code's plugin system checks for ~/.claude/.i-have-adhd-always at session initialization. This file-based flag avoids modifying global Claude settings while allowing per-user, persistent activation. Delete the file to disable auto-loading without uninstalling the plugin.
How do I verify the skill is active in my current session?
Check that responses follow the 10-rule format: action-first opening, numbered steps, progress restatement ("step X of Y"), concrete time units, no preamble, and capped lists. If rules aren't applying, re-run the activation command (/i-have-adhd) and confirm the plugin installed correctly per [AGENTS.md](https://github.com/ayghri/i-have-adhd/blob/main/AGENTS.md).
Have a question about this repo?
These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:
curl -s "https://instagit.com/install.md" Maintain an open-source project? Get it listed too →