# How to Deploy the i-have-adhd Application: Complete Agent Skill Installation Guide > Easily deploy i-have-adhd application agent skill into AI coding assistants. Follow this guide for simple installation with no build steps or server config needed. - Repository: [Ayoub Ghriss/i-have-adhd](https://github.com/ayghri/i-have-adhd) - Tags: how-to-guide - Published: 2026-07-30 --- **Deploying the i-have-adhd application involves installing the agent skill into compatible AI coding assistants through marketplace commands or manual file copies, requiring no build steps, server configuration, or runtime dependencies.** The i-have-adhd project by ayghri is an **Agent Skill** that enhances AI coding assistants with structured output rules and the `/i-have-adhd` slash command. Because the repository contains only declarative files—specifically [`SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/SKILL.md), [`plugin.json`](https://github.com/ayghri/i-have-adhd/blob/main/plugin.json), and agent-specific configuration files under `skills/i-have-adhd/`—deploying the application is essentially a copy operation rather than a traditional software deployment. ## Understanding the Architecture Before deploying, recognize that i-have-adhd is not a standalone server application but a **rule-based skill** consumed by AI agents. The repository structure reflects this: - **[`skills/i-have-adhd/SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md)**: The core rule definition that sets `disable-model-invocation: true` and defines the `/i-have-adhd` command behavior - **[`skills/i-have-adhd/plugin.json`](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/plugin.json)**: Minimal metadata describing the plugin name and purpose for marketplace-aware agents - **`skills/i-have-adhd/agents/`**: Contains agent-specific loaders like [`openai.yaml`](https://github.com/ayghri/i-have-adhd/blob/main/openai.yaml) and [`gemini.toml`](https://github.com/ayghri/i-have-adhd/blob/main/gemini.toml) that translate the skill into each platform's native format Since the skill ships as pure data with no binaries or server-side components, deployment completes once the agent recognizes and registers the files. ## Prerequisites You need a supported AI coding assistant. The repository explicitly supports: - Claude Code - Codex CLI - Zed Editor - Gemini CLI - Hermes - Antigravity (`agy`) - OpenCode and Cursor (via generic skill loaders) ## Deployment Methods by Agent Each platform provides distinct installation mechanisms, though all ultimately pull the same files from `github.com/ayghri/i-have-adhd`. ### Claude Code (Marketplace) Claude Code supports direct marketplace installation and optional "always-on" configuration: ```bash # Install from marketplace claude plugin marketplace add ayghri/i-have-adhd claude plugin install i-have-adhd@i-have-adhd # Verify installation claude plugin list # Enable always-on mode (optional) touch ~/.claude/.i-have-adhd-always ``` ### Codex CLI Similar to Claude Code, Codex uses a marketplace workflow: ```bash # Add marketplace source and install codex plugin marketplace add ayghri/i-have-adhd --ref main codex plugin add i-have-adhd@i-have-adhd # Verify codex plugin list ``` ### Zed Editor Zed requires manual file placement since it reads skills from a local directory: ```bash # Clone and copy to Zed's skills directory git clone https://github.com/ayghri/i-have-adhd cp -R i-have-adhd/skills/i-have-adhd ~/.config/zed/skills/ ``` ### Gemini CLI For Gemini CLI, you copy the agent-specific configuration file directly: ```bash mkdir -p ~/.gemini/commands 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 ``` ### Hermes Hermes provides a dedicated skills installer: ```bash hermes skills install ayghri/i-have-adhd/skills/i-have-adhd ``` ### Antigravity (agy) The `agy` CLI handles the repository directly: ```bash agy plugin install https://github.com/ayghri/i-have-adhd ``` ### OpenCode and Cursor These editors use a generic npm-based skill loader: ```bash # Workspace-local installation npx skills add ayghri/i-have-adhd # Global installation npx skills add ayghri/i-have-adhd -g ``` ## Post-Deployment Configuration **Important**: The skill is **disabled by default** as specified in [`SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/SKILL.md) (`disable-model-invocation: true`). This means the agent will not automatically apply the i-have-adhd rules until you either: 1. Type the `/i-have-adhd` command explicitly in your session, or 2. Create the always-on flag file (e.g., `~/.claude/.i-have-adhd-always` for Claude Code) Updates are handled through each agent's native update mechanism—simply re-run the installation command or use the agent's plugin update CLI to fetch newer rule changes from the repository. ## Summary - **The i-have-adhd application is an Agent Skill**, not a server application, consisting of declarative files in `skills/i-have-adhd/` - **Deployment requires no compilation**—simply copy files or use marketplace commands specific to your AI assistant - **Key files** include [`SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/SKILL.md) for rules, [`plugin.json`](https://github.com/ayghri/i-have-adhd/blob/main/plugin.json) for metadata, and files in `agents/` for platform-specific loading - **The skill is disabled by default** and activates only when you run `/i-have-adhd` or enable always-on mode - **Supported platforms** include Claude Code, Codex, Zed, Gemini CLI, Hermes, and others, each with specific one-line installation commands ## Frequently Asked Questions ### Do I need to build or compile anything to deploy i-have-adhd? No. According to the repository structure in `ayghri/i-have-adhd`, the application contains only text files like [`SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/SKILL.md) and [`plugin.json`](https://github.com/ayghri/i-have-adhd/blob/main/plugin.json). There are no binaries to compile, no Docker containers to run, and no dependencies to install. Deployment is a file copy operation. ### Why isn't the skill working immediately after installation? The skill explicitly sets `disable-model-invocation: true` in [`skills/i-have-adhd/SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md). This design disables automatic model invocation, meaning the rules only apply when you type the `/i-have-adhd` command or create an "always-on" flag file in your agent's configuration directory. ### How do I update the skill when the repository changes? Run the same installation command again or use your agent's native update mechanism. For example, in Claude Code you would re-run `claude plugin install i-have-adhd@i-have-adhd`, while for Zed you would pull the latest repository changes and recopy the `skills/i-have-adhd/` directory. ### Can I use i-have-adhd with multiple AI assistants simultaneously? Yes. Because the repository provides agent-specific configurations in `skills/i-have-adhd/agents/` (including [`openai.yaml`](https://github.com/ayghri/i-have-adhd/blob/main/openai.yaml) and [`gemini.toml`](https://github.com/ayghri/i-have-adhd/blob/main/gemini.toml)), you can deploy the same skill to Claude Code, Gemini CLI, and others on the same machine. Each assistant maintains its own copy of the configuration files.