Forbidden Openers, Recaps, and Closers in i-have-adhd: Complete Style Guide

The i-have-adhd skill strictly prohibits conversational padding—including openers like "Great question," recaps such as "I've now done X, Y, and Z," and closers like "Hope this helps"—to ensure every response delivers immediate, actionable value for readers with ADHD.

The ayghri/i-have-adhd repository defines a specialized conversational skill designed to respect the cognitive needs of users with ADHD. According to the source code in skills/i-have-adhd/SKILL.md, the skill enforces rule 10 ("No preamble, no recap, no closing pleasantries") by explicitly forbidding specific categories of phrasing that dilute focus or delay actionable content.

What Are Forbidden Openers, Recaps, and Closers in i-have-adhd?

The skill categorizes prohibited conversational fluff into three groups, each defined in specific lines of the SKILL.md file.

Forbidden Openers

The skill bans any preamble that delays the core answer. As documented in lines 111-113 of skills/i-have-adhd/SKILL.md, specifically prohibited phrases include:

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

Forbidden Recaps

After completing steps, the skill forbids summarizing what was just accomplished. The disallowed pattern defined at line 113 is:

• "I’ve now done X, Y, and Z, which means…"

Forbidden Closers

Polite sign-offs that invite further clarification without concrete next steps are banned. Line 115 of SKILL.md explicitly prohibits:

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

Why These Phrases Are Prohibited

Rule 10 exists to maintain cognitive momentum. For readers with ADHD, conversational fluff creates friction and distracts from task completion. By eliminating openers, the response starts with the command or information needed right now. Removing recaps prevents redundant mental processing of completed steps. Deleting closers avoids vague social loops and instead forces the assistant to specify the exact next concrete action.

Correct vs. Incorrect Usage Examples

Compliant Response Structure

When the i-have-adhd skill is active, structure responses to open immediately with the action, provide the body content, and close with a specific next step:

Run npm install jsonwebtoken, then edit src/auth.ts:42.

1. Open src/auth.ts
2. Replace verifyToken with the snippet below
3. Run npm test -- auth.spec.ts

Next: run npm test and paste the first failing line.

Common Violations to Avoid

Incorrect opener:

Let me explain how to fix this.

Incorrect recap:

I've now done the install, the edit, and the test, which means everything should work.

Incorrect closer:

Hope this helps! Let me know if you need anything else.

Summary

• Forbidden openers in i-have-adhd include "Great question," "Let me…", and "To answer your question…" as defined in skills/i-have-adhd/SKILL.md lines 111-113.
• Forbidden recaps such as "I've now done X, Y, and Z" are banned to prevent redundant summarization after task completion.
• Forbidden closers like "Hope this helps" and "Let me know if you need anything else" are prohibited per line 115 to eliminate vague conversational exits.
• Rule 10 requires starting immediately with actionable content, omitting mid-conversation summaries, and ending with a specific next step rather than a polite offer of further help.

Frequently Asked Questions

What is the purpose of forbidding openers in i-have-adhd?

The prohibition eliminates preamble that delays actionable information. As implemented in ayghri/i-have-adhd, removing phrases like "Great question" ensures the response begins immediately with the command, code, or instruction the user needs, reducing cognitive load for ADHD readers.

Are there any exceptions to the forbidden recaps rule?

No. According to the skill definition in skills/i-have-adhd/SKILL.md line 113, recaps such as "I've now done X, Y, and Z, which means…" are universally disallowed. The assistant must proceed to the next concrete step without summarizing completed work.

How should I end a response if I can't use "Let me know if you need anything else"?

Replace vague closers with a specific next action. Instead of offering general help, state the exact step the user should take next, such as "Next: run npm test and paste the first failing line" or "Deploy the changes with git push origin main."

Where are these conversational rules defined in the repository?

All forbidden openers, recaps, and closers are documented in rule 10 of skills/i-have-adhd/SKILL.md at lines 111-115, which specifies the "No preamble, no recap, no closing pleasantries" constraint.

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