# Banned Openers and Closers in the i‑have‑adhd Skill: A Complete Guide

> Discover which openers and closers are banned in the i-have-adhd skill. Learn how this enforces action-first communication for better focus and memory.

- Repository: [Ayoub Ghriss/i-have-adhd](https://github.com/ayghri/i-have-adhd)
- Tags: api-reference
- Published: 2026-09-02

---

**The *i‑have‑adhd* skill bans all pre‑amble openers (e.g., "Great question," "Let me …") and polite closers (e.g., "Hope this helps," "Let me know if you need anything else") to enforce action‑first communication that respects limited working memory.**

The [ayghri/i‑have‑adhd](https://github.com/ayghri/i-have-adhd) repository provides a specialized skill designed for users with ADHD. Its core principle is **radical brevity**: every response must start with a concrete, runnable step and end with a single, concrete next action. Anything else—especially conversational padding—gets stripped before delivery.

## Banned Openers: No Preambles Allowed

Rule 10 in [[`SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/SKILL.md) line 109](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md#L109) explicitly forbids introductory phrases that delay actionable content. The skill treats these as "preambles" and removes them entirely.

**Common banned openers include:**

- "Great question,"
- "Let me …"
- "I’ll …"
- "Sure!"
- "Looking at your …"
- "To answer your question …"

These phrases consume cognitive load without advancing the task. The skill's parser identifies and strips them so the first line seen by the user is always executable.

## Banned Closers: No Polite Sign‑offs

Similarly, [[`SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/SKILL.md) line 113](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md#L113) prohibits closing pleasantries that don't specify a concrete next step.

**Prohibited closers include:**

- "Let me know if you need anything else,"
- "Hope this helps,"
- "Happy to clarify,"
- "Feel free to ask."

The skill removes these because they leave the user hanging rather than directing them toward completion.

## Before and After: What Gets Changed

Here's how the transformation works in practice.

**Banned style (stripped by the skill):**

```

Great question, here’s what you need to do.
1. Open src/app.js
2. Add the handler
Hope this helps.

```

**Allowed style (preserved by the skill):**

```

Open `src/app.js` and add the handler.
1. Open src/app.js
2. Insert the new function
Next: run `npm test` to verify.

```

Notice the difference: the revised version opens with a direct command and closes with a specific, labeled next action ("Next:"). No wasted words, no decision fatigue.

## Implementation Details

The banned openers and closers are defined in the skill's configuration files. According to the source code, the enforcement happens at the response generation layer before any text reaches the user interface.

Key files in the repository:

- **[`skills/i-have-adhd/SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md)** — Contains Rule 10 with the authoritative list of forbidden phrases and the reasoning behind each category
- **[`README.md`](https://github.com/ayghri/i-have-adhd/blob/main/README.md)** — Overview of the skill's design philosophy and integration patterns
- **[`AGENTS.md`](https://github.com/ayghri/i-have-adhd/blob/main/AGENTS.md)** — Technical specification for runtime adapters that implement the stripping logic

## Summary

- **Banned openers** ("Great question," "Let me …", etc.) are preambles that delay actionable content
- **Banned closers** ("Hope this helps," "Feel free to ask," etc.) are sign‑offs that fail to specify next steps
- Both categories are defined in [[`SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/SKILL.md)](https://github.com/ayghri/i-have-adhd/blob/main/skills/i-have-adhd/SKILL.md) Rule 10 (lines 109 and 113)
- The skill preserves only content that starts with a concrete action and ends with a labeled next step

## Frequently Asked Questions

### Why does the i‑have‑adhd skill ban common polite phrases?

Polite openers and closers create **cognitive overhead** without advancing the user's task. For people with ADHD, this padding competes for limited working memory and can trigger avoidance behaviors. The skill removes these friction points entirely.

### Can I customize which openers and closers get banned?

The current implementation uses a fixed list in [`SKILL.md`](https://github.com/ayghri/i-have-adhd/blob/main/SKILL.md). Runtime adapters built from [`AGENTS.md`](https://github.com/ayghri/i-have-adhd/blob/main/AGENTS.md) specifications implement the stripping logic, so modifications would require changes to both the rule definitions and the adapter code.

### What replaces the banned closers?

The skill requires a **labeled next action** instead—typically prefixed with "Next:" followed by a single, concrete step. This provides closure while maintaining momentum toward task completion.

### Does this skill affect how I should write prompts, or only how responses are formatted?

The restrictions apply to **response generation**. However, understanding the banned openers and closers helps you write prompts that align with the skill's expectations, since it will strip similar padding from its own outputs regardless of input style.