How to Configure Environment Variables for i-have-adhd Development

To configure environment variables for i-have-adhd development, create a .env file in the repository root with your LLM API keys, then restart your Claude Code or Codex session to load the values.

The ayghri/i-have-adhd repository provides a plugin skill designed to run inside AI agents such as Claude Code, Codex, or Gemini-enabled environments. When you configure environment variables for i-have-adhd development, you enable the skill to authenticate with OpenAI, Anthropic, and Google APIs. This setup follows the standard dotenv pattern and requires no code changes to the core logic.

Creating the .env File

Place a file named .env in the repository root—next to README.md and plugin.json. According to the installation flow documented in README.md (lines 24-31 for Claude Code and lines 36-44 for Codex), the runtime expects secrets to reside in this local environment file.

Never commit this file. The repository’s .gitignore already excludes .env by default to prevent accidental exposure of credentials.

Required API Keys

The skill reads several variables at startup to route requests to the appropriate LLM providers. Define only the keys for the providers you intend to use:

• OPENAI_API_KEY – Required for GPT-4 or GPT-3.5-turbo calls. Format: sk-xxxxxxxxxxxxxxxxxxxx
• ANTHROPIC_API_KEY (or CLAUDE_API_KEY) – Required for Claude model access. Format: sk-xxxxxxxxxxxxxxxxxxxx
• GEMINI_API_KEY – Required for Google Gemini integration. Format: AIzaxxxxxxxxxxxxxxxxxxxx
• PLUGIN_API_KEY – Optional. Use this if you publish the plugin to a marketplace that requires authentication.

How the Runtime Loads Configuration

The plugin code automatically loads these variables using the popular dotenv pattern. As implemented in ayghri/i-have-adhd, the runtime reads the .env file at startup before initializing the skill logic defined in skills/i-have-adhd/SKILL.md. No manual load_dotenv() calls are required in your implementation.

Example .env Configuration

# .env – Place in repository root (same level as README.md)

OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxx
ANTHROPIC_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxx
GEMINI_API_KEY=AIzaxxxxxxxxxxxxxxxxxxxxxxxxxxxx

After saving the file, restart your agent session. For Claude Code, reload the /i-have-adhd command; for Codex, restart the $i-have-adhd session. The skill will pick up the new values immediately.

Summary

• Create a .env file in the repository root next to README.md and plugin.json
• Populate it with OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, or CLAUDE_API_KEY as needed
• Verify .env is listed in .gitignore before adding secrets
• Restart your Claude Code or Codex session to apply changes

Frequently Asked Questions

What file should I edit to add API keys for i-have-adhd development?

Create a .env file in the repository root. The runtime looks for this file at startup to load secrets via the dotenv pattern, as referenced in the installation instructions in README.md and INSTALL.md.

Is it safe to commit the .env file to Git?

No. The repository includes .env in its .gitignore by default. Committing this file would expose your OPENAI_API_KEY, ANTHROPIC_API_KEY, and other secrets publicly.

Do I need to provide all three API keys to run the plugin?

No. Provide only the keys for the LLM providers you plan to use. For example, if you only use Claude, set ANTHROPIC_API_KEY and omit the others. The skill will use whichever credentials are present.

Where does the plugin read the environment variables from?

The skill reads variables from the .env file in the repository root using the dotenv pattern. This behavior is implied by the local installation flows documented in README.md lines 24-31 (Claude Code) and lines 36-44 (Codex), where the runtime loads the local environment before executing the skill logic defined in skills/i-have-adhd/SKILL.md.

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