Introductory Phrases and Closers That i-have-adhd Explicitly Blocks

The i-have-adhd skill prohibits conversational openers like "Great question" and closers like "Hope this helps" to ensure every response is immediately actionable for users with ADHD.

The ayghri/i-have-adhd repository defines an AI interaction skill optimized for attention deficit hyperactivity disorder (ADHD) cognitive patterns. As part of its strict formatting protocol, the skill explicitly forbids specific introductory phrases and closers that introduce cognitive overhead without contributing concrete next steps. These constraints are centrally defined in the skill’s canonical definition file to maintain consistent, fluff-free communication.

Forbidden Introductory Phrases and Closers in SKILL.md

The constraints are formally defined in skills/i-have-adhd/SKILL.md at lines 107-113, which enumerate specific introductory phrases and closers banned under the "No preamble" rule. This rule applies throughout the session, ensuring zero conversational overhead.

Disallowed Opening Phrases

According to the source code analysis, the following introductory phrases (openers) are forbidden because they delay actionable content:

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

These phrases are classified as preamble that does not advance the user’s task.

Disallowed Closing Phrases

Similarly, the skill forbids closing pleasantries and recap statements that fail to specify the next step:

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

The rule requires that the final line of every response end with the next actionable step rather than a social nicety.

Practical Examples: Disallowed vs Allowed Responses

The following examples demonstrate how the skill transforms conversational AI patterns into ADHD-optimized formats:

Opening a reply:

Disallowed: "Let me explain how to fix this."
Allowed: Run npm install jsonwebtoken, then edit src/auth.ts:42.

Closing a reply:

Disallowed: "Hope this helps! Let me know if you need anything else."
Allowed: Next: run npm test and paste the first failing line.

Mid-response preamble:

Disallowed: "Sure! Here’s what you asked for."
Allowed: [No preamble – start directly with the actionable step]

Ending with a pleasantry:

Disallowed: "Feel free to ask any follow-up questions."
Allowed: Next: open the file and verify the change.

Implementation Architecture

These restrictions are enforced through the canonical skill definition file. In skills/i-have-adhd/SKILL.md, the forbidden phrases are explicitly listed alongside the architectural requirement that every response must begin with a concrete action and conclude with the subsequent step. The implementation references lines 107-113, which define the "No preamble" rule governing both openers and closers throughout the session.

Summary

• Introductory phrases and closers such as "Great question" and "Hope this helps" are explicitly forbidden in the i-have-adhd skill.
• The restrictions are codified in skills/i-have-adhd/SKILL.md at lines 107-113 under the "No preamble" rule.
• Responses must open with immediate technical action rather than conversational preamble.
• Every response must close with the next concrete step instead of social pleasantries.
• These constraints reduce cognitive load for ADHD users by eliminating non-actionable conversational fluff.

Frequently Asked Questions

What specific introductory phrases does i-have-adhd prohibit?

The skill bans openers including "Great question," "Let me …," "I’ll …," "Sure!," "Looking at your …," and "To answer your question …" These are defined in the "No preamble" section of SKILL.md as phrases that delay actionable content.

Where are the forbidden phrase rules defined in the codebase?

The prohibited openers and closers are specified in skills/i-have-adhd/SKILL.md at lines 107-113. This file serves as the canonical skill definition that governs response formatting throughout the session.

Why does i-have-adhd eliminate closing pleasantries like "Hope this helps"?

Closing pleasantries are removed because they do not contribute concrete next steps for the user. The skill is architected for ADHD cognitive patterns, requiring that every line—including the final line—advances the task or specifies the immediate next action.

How should responses start and end if not with traditional greetings?

Responses must start with an immediate actionable instruction, such as a command or code edit. They must end with a specific next step, formatted as "Next: [action]" or equivalent concrete direction, ensuring zero conversational overhead between the user and their goal.

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