How i-have-adhd Numbered Lists Handle Multi-Step Tasks

The i-have-adhd skill enforces a strict five-item cap on numbered lists, requiring each step to represent a single bounded action under two minutes, formatted inside fenced code blocks with progress restatement after completion.

The ayghri/i-have-adhd repository defines a specialized skill for structuring assistant outputs to accommodate cognitive accessibility needs. When handling i-have-adhd numbered lists for complex procedures, the system applies rigid constraints defined in skills/i-have-adhd/SKILL.md to minimize working memory load and decision paralysis.

The Five-Item Cognitive Cap

According to Rule 9 in skills/i-have-adhd/SKILL.md (lines 105-108), the system caps all numbered lists at five items to maintain cognitive manageability. When procedures exceed this limit, the skill splits them into immediate "do now" steps versus deferred "later" groups, preventing overwhelm.

Bounded and Minimal Actions

Single-Action Steps

Rule 2 (lines 44-48) mandates that each list item must represent a single, bounded action completable in under two minutes. The specification explicitly prohibits compound "and then" sequences within individual steps, ensuring every item is immediately actionable without hidden sub-tasks.

Step Minimization

The same rule (lines 48-50) requires folding trivial actions into preceding steps and eliminating unnecessary intermediaries. The list must contain the fewest steps required to achieve the goal, stripping away preparatory fluff or tangential suggestions.

Formatting and State Management

Visual Separation Requirements

Rule 2 (lines 52-57) dictates that i-have-adhd numbered lists appear inside fenced code blocks for clear visual separation. The list must precede any explanatory prose, ensuring users encounter actionable instructions before contextual details that might distract from execution.

Progress Restatement Protocol

Per Rule 5 (lines 75-80), the assistant must restate current progress after completing each step (e.g., "Step 2 of 3 done") before presenting the next numbered item. This provides explicit state tracking for users with limited working memory, reducing the cognitive overhead of remembering position within a sequence.

Before and After: Formatting Examples

Unstructured responses create cognitive drag by mixing context with commands. Here is how the skill transforms typical assistant output:

Before (unstructured prose):

> Great question! Let me think about this. Your auth flow has a few moving pieces: the middleware, the token verification, and the cookie handling. Looking at `src/auth.ts`, the `verifyToken` function (around lines 42-58) seems to be using an older `jsonwebtoken` API. One approach would be to update the package and rewrite that function. After making the change, you'd want to run the auth tests to confirm nothing breaks. By the way, you might also want to look at your dependency versions overall. Hope this helps! Let me know if you want to dig deeper.

After (structured numbered list):

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

Next: run the tests and paste the first failing line.

Five-Item Cap Demonstration:

1. Install the latest `jsonwebtoken` package.  
2. Open `src/auth.ts`.  
3. Update the `verifyToken` implementation.  
4. Run the unit tests.  
5. Verify the login flow works locally.  

Later: consider updating related documentation.  

Source Implementation

The constraints are codified primarily in skills/i-have-adhd/SKILL.md, which defines Rule 2 for numbering multi-step tasks and Rule 9 for the cognitive cap. Supporting documentation appears in README.md with concrete before/after examples, while plugin.json registers the skill within the coding assistant platform.

Summary

• i-have-adhd numbered lists are strictly capped at five items to respect working memory limitations.
• Each step must represent a single, bounded action completable in under two minutes with no compound sequences.
• Lists render inside fenced code blocks and precede all explanatory text to prioritize action over context.
• The assistant must restate progress after each completed step to provide explicit state tracking.
• Long procedures split into "do now" and "later" groups rather than exceeding the five-item limit.

Frequently Asked Questions

Why does i-have-adhd limit numbered lists to five items?

The five-item cap defined in Rule 9 (lines 105-108 of skills/i-have-adhd/SKILL.md) prevents cognitive overload. Research on working memory suggests most individuals can hold approximately four to seven items in mind simultaneously; capping at five ensures users with ADHD can track progress without exceeding cognitive bandwidth.

How does the skill handle procedures requiring more than five steps?

When procedures exceed five steps, the skill splits them into immediate "do now" versus deferred "later" groups. The first list contains the urgent five actions, followed by a separate "Later:" section indicating subsequent steps, allowing users to focus exclusively on the current subset.

What is the "state restatement" requirement in i-have-adhd?

Rule 5 (lines 75-80) requires the assistant to explicitly restate progress after completing each step, such as declaring "Step 2 of 3 done." This externalizes working memory by making the current position within the sequence explicit, reducing the user's burden of tracking their own progress.

Where are the numbered list formatting rules defined?

The numbered list constraints reside in skills/i-have-adhd/SKILL.md under Rule 2 for multi-step task structure and Rule 9 for the cognitive cap. Additional implementation context appears in the repository's README.md, with plugin registration handled in plugin.json.

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