Rule 10: No Preamble, No Recap, No Closing Pleasantries Explained (i-have-adhd)

Rule 10 in the i-have-adhd skill forbids all conversational padding—including opening acknowledgments, post-task summaries, and sign-off phrases—to ensure every response begins with an actionable item and ends with only the next concrete step.

The open-source repository ayghri/i-have-adhd provides a constraint-based communication framework designed to eliminate cognitive overhead for users with ADHD. Rule 10 serves as the final gatekeeper in the ten-rule skill set, stripping away social lubricant language that often obscures critical instructions in AI-assisted workflows.

The Three Prohibitions of Rule 10

According to skills/i-have-adhd/SKILL.md (lines 109‑115), Rule 10 enforces a strict ban on three categories of conversational filler that persist for the entire session duration.

Forbidden Preambles

Opening phrases that announce intent or acknowledge the user’s question are prohibited. The source code explicitly lists these forbidden openers:

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

These phrases delay the delivery of actual content and force the reader to scan past unnecessary politeness markers.

Forbidden Recaps

Summary sentences that repeat completed actions after a task is finished are banned. The skill specifically forbids constructions like:

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

This rule prevents redundant cognitive loops where the AI restates work already visible in the chat history, respecting the user’s working memory limitations.

Forbidden Closing Pleasantries

Sign-off phrases that invite further interaction or express goodwill are stripped out. Forbidden closers include:

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

Instead, responses must terminate with a concrete next action or a specific file path.

Source Definition and Enforcement

The rule is hardcoded in the skill definition file with zero tolerance for deviation. As implemented in ayghri/i-have-adhd, the constraints remain active for the duration of the session unless the user explicitly issues the deactivation command.

To lift Rule 10, the user must state "stop adhd mode" or "normal mode". Until then, every generated response undergoes implicit filtering to remove the three prohibited categories.

Practical Examples: Correct vs. Incorrect Usage

The repository provides concrete examples demonstrating compliant and non-compliant outputs.

Correct usage (adheres to Rule 10):

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

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

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

Incorrect usage (violates Rule 10):

"Great question! Let me think about this. You should run `npm install jsonwebtoken`… After you’ve done that, hope this helps."

The incorrect example violates all three prohibitions: it opens with an acknowledgment ("Great question"), contains mid-process meta-commentary ("Let me think about this"), and closes with a pleasantry ("hope this helps").

Why Rule 10 Matters for ADHD Workflows

Neurodivergent users often experience decision paralysis when faced with ambiguous social framing or redundant information. By mandating that the first line of every response contain the immediately actionable item and the last line specify the next concrete step, Rule 10 reduces cognitive load to the absolute minimum.

The rule interacts with the Persistence system documented in SKILL.md: once activated, these constraints apply to every subsequent turn in the conversation. This consistency eliminates the anxiety of parsing varying politeness levels or hunting for buried instructions.

Summary

• Rule 10 prohibits preambles, recaps, and closing pleasantries in every response while the skill is active.
• Forbidden phrases are explicitly listed in skills/i-have-adhd/SKILL.md lines 109‑115, including "Great question," "Let me know if you need anything else," and "I’ve now done X, Y, and Z."
• Compliant responses start with an actionable command and end with a specific next step, with no conversational wrapper.
• The constraints persist for the entire session until the user explicitly requests "stop adhd mode" or "normal mode."

Frequently Asked Questions

What specific phrases does Rule 10 ban?

Rule 10 prohibits common conversational openers like "Great question" and "Looking at your…," post-task summaries such as "I’ve now done X, Y, and Z," and closers including "Hope this helps" and "Happy to clarify." The complete list is codified in the SKILL.md definition file.

How do I turn off Rule 10 constraints?

You must explicitly state "stop adhd mode" or "normal mode" to deactivate Rule 10 and restore default conversational behavior. Until then, the prohibitions remain enforced for every response in the active session.

Why does the i-have-adhd skill forbid polite closing phrases?

Closing pleasantries like "Let me know if you need anything else" introduce ambiguity about whether the task is complete. For ADHD users, this certainty gap can trigger distraction or rumination. Rule 10 replaces ambiguity with a concrete next step, preserving executive function bandwidth.

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