What Types of Language Does i-have-adhd Avoid? Preambles, Recaps, and Closers Explained

The i-have-adhd skill explicitly bans preambles, recaps, and closers to eliminate cognitive overhead and deliver immediate, actionable responses.

The open-source ayghri/i-have-adhd repository defines a structured communication framework designed for readers with ADHD. Central to its design is Rule 10, which codifies exactly what types of language the skill avoids to reduce working-memory load and prevent decision paralysis.

The Three Forbidden Language Patterns

Preambles

Preambles are opening phrases such as "Great question," "Let me…," "Sure!," or "Looking at your…". These add unnecessary contextual noise that the reader must mentally filter out before reaching actionable content. By stripping these openers, the skill ensures the first line of every response contains immediate value.

Recaps

Recaps are post-task summaries that restate completed actions (e.g., "I’ve now done X, Y, and Z, which means…"). These summaries duplicate information the reader already sees and pull focus away from the next concrete step. The skill eliminates recaps to maintain forward momentum and reduce redundant cognitive processing.

Closers

Closers include polite sign-offs like "Hope this helps," "Let me know if you need anything else," "Happy to clarify," or "Feel free to ask." These phrases create a false sense of completion that encourages the reader to wait for confirmation rather than proceeding immediately with the workflow.

Rule 10 in the Source Code

These prohibitions are hard-coded in skills/i-have-adhd/SKILL.md at lines 107-113:

"Forbidden openers: 'Great question,' 'Let me…', 'I’ll…', 'Sure!', 'Looking at your…', 'To answer your question…'"
"Forbidden recaps after a completed task: 'I’ve now done X, Y, and Z, which means…'"
"Forbidden closers: 'Let me know if you need anything else,' 'Hope this helps,' 'Happy to clarify,' 'Feel free to ask.'"

Before and After: Response Comparison

Non-Compliant Response (Contains Filler Language)

Great question! Let me think about this. Your authentication flow has a few moving pieces. First, open `src/auth.ts`, then locate the `verifyToken` function. After updating it, run the tests.

I’ve now edited the file and the tests pass, which means the issue is resolved.

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

Compliant Response (Action-First Style)

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

1. Open `src/auth.ts`.
2. Replace `verifyToken` (lines 42-58) with the new snippet.
3. Run `npm test -- auth.spec.ts`.

Next: Paste the first failing line if any test fails.

The compliant version leads with the executable command, uses numbered steps for scannability, and ends with a concrete next action rather than a social closer. This structure aligns with the dopamine-friendly, working-memory-conscious goals of the skill.

Summary

• Preambles are eliminated because opening pleasantries force readers to filter out noise before accessing useful content.
• Recaps are prohibited as they restate completed work and distract from the next immediate step.
• Closers are banned to prevent false completion signals that interrupt workflow momentum.
• These rules are enforced in skills/i-have-adhd/SKILL.md to ensure consistent, ADHD-optimized output across all assistant interactions.

Frequently Asked Questions

What exactly counts as a preamble in i-have-adhd?

Phrases like "Great question," "Let me...", "Sure!", and "Looking at your..." are explicitly classified as preambles. According to the source code in SKILL.md, these openers add unnecessary context that increases cognitive load before delivering the actual answer.

Why does i-have-adhd avoid recaps after completing tasks?

Recaps such as "I’ve now done X, Y, and Z..." restate information the user already possesses, which wastes working memory and delays the next actionable instruction. The skill prioritizes forward momentum over retrospective summaries.

Are polite closers like "Hope this helps" completely banned?

Yes, closers including "Hope this helps," "Happy to clarify," and "Feel free to ask" are prohibited under Rule 10. These phrases generate a false sense of finality that encourages the reader to pause and wait rather than proceeding immediately with their workflow.

Where are these language rules documented in the repository?

The canonical definitions reside in skills/i-have-adhd/SKILL.md at lines 107-113. This file forms the core specification for the skill's output style, guiding developers on how to implement ADHD-friendly communication patterns.

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