Rule 3 in i-have-adhd: End with One Concrete Next Action Explained

Rule 3 ensures every response terminates with a single, immediately executable task under two minutes to counteract ADHD-related working memory overload and action initiation friction.

The i-have-adhd skill structures AI outputs to accommodate neurodivergent cognitive patterns. Rule 3 (End with one concrete next action) is a mandatory constraint defined in skills/i-have-adhd/SKILL.md that eliminates ambiguous closings and replaces them with precise, bite-sized directives.

The Purpose of Rule 3 in i-have-adhd

Rule 3 serves as a behavioral safeguard against three specific ADHD challenges outlined in the skill’s "Five facts" description (§2.1-2.5). By mandating a concrete final step, the rule transforms passive information consumption into active momentum.

Closing Information Loops

ADHD often involves working-memory overload, where readers struggle to retain multi-step sequences in mind. Rule 3 forces the closure of any open information loop by appending a single, actionable item at the end of every response. As implemented in skills/i-have-adhd/SKILL.md, this ensures the reader knows exactly what to execute next without mentally replaying earlier instructions.

Reducing Decision Paralysis

Presenting multiple options or vague offers of further help triggers action initiation friction. Rule 3 eliminates the "too many options" effect by restricting the closing to exactly one concrete action. This constraint removes the cognitive burden of prioritization, allowing the reader to initiate movement immediately rather than stalling in analysis.

Creating Immediate Dopamine Rewards

The "under two minutes" constraint guarantees the prescribed task is small enough to complete before distraction intervenes. According to the source code analysis, this generates a quick win that produces immediate dopamine, reinforcing the reward loop and encouraging continued engagement with the workflow.

Rule 3 Formal Definition in SKILL.md

The rule is canonically defined in skills/i-have-adhd/SKILL.md at lines 57-60:

"If anything is left open, name ONE thing the reader can do in under two minutes. Even open the file counts."

This specification appears in the section contrasting acceptable versus prohibited response patterns. The enforcement mechanism is straightforward: if a response does not terminate with a "Next:" directive followed by a specific, timed action, it violates the skill protocol.

Bad vs. Good Examples from the Source

The SKILL.md file at lines 61-63 provides explicit contrast examples demonstrating the rule’s application:

Bad (violates Rule 3):

Hope that helps. Let me know if you want to dig deeper.

Good (complies with Rule 3):

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

The good example names a specific command (npm test) and a concrete deliverable (the first failing line), satisfying the under-two-minute constraint. The bad example offers an open-ended invitation that requires the reader to formulate their own next step.

Implementing Rule 3 in Practice

When constructing responses under this skill, you must append the concrete action after any explanatory prose or numbered sequence. The following examples from the repository demonstrate proper formatting:

Correct structure with numbered steps:

1. Open `src/auth.ts`.
2. Replace the `verifyToken` function with the snippet below.
3. Run the test suite.

**Next:** Execute `npm test -- auth.spec.ts` and share the first failure you see.

Correct structure for simple queries:

… (explanation omitted)

**Next:** Run `npm test -- auth.spec.ts` and paste the first failing line here.

Incorrect structure (missing concrete action):

… (steps omitted)

Let me know if you want me to help with anything else.

The AGENTS.md file maps this skill to runtime adapters, ensuring that all agent outputs enforce the Rule 3 constraint across the project. The README.md provides high-level context for how this behavior integrates into the broader i-have-adhd methodology.

Summary

• Rule 3 mandates that every response end with exactly one concrete next action to mitigate ADHD-specific cognitive challenges.
• The action must be completable in under two minutes, ensuring dopamine-friendly quick wins.
• The rule is formally defined in skills/i-have-adhd/SKILL.md (lines 57-60) and enforced via AGENTS.md.
• Valid examples use a "Next:" prefix followed by specific commands or steps; invalid examples use open-ended offers of future assistance.
• This pattern combats working-memory overload and decision paralysis by closing information loops and removing option ambiguity.

Frequently Asked Questions

What is the maximum duration for a Rule 3 next action?

The SKILL.md specification explicitly limits Rule 3 actions to tasks completable in under two minutes. This time constraint ensures the task generates immediate dopamine before distraction or fatigue can interrupt the workflow.

Where is Rule 3 formally defined in the i-have-adhd repository?

Rule 3 is formally defined in skills/i-have-adhd/SKILL.md at lines 57-60, with contrast examples showing prohibited versus compliant patterns at lines 61-63. The enforcement architecture is referenced in AGENTS.md.

Why does Rule 3 specify only ONE action instead of multiple options?

Restricting the closing to a single action eliminates decision paralysis, a common ADHD challenge where multiple options stall initiation. By prescribing exactly one concrete step, Rule 3 removes the cognitive overhead of prioritization and forces immediate movement.

How does Rule 3 interact with other rules in the i-have-adhd skill?

Rule 3 functions as an output-shaping constraint that operates alongside other structural rules in the skill. While the "Five facts" (§2.1-2.5) establish the cognitive rationale for the approach, Rule 3 specifically governs response termination to ensure every interaction ends with an executable directive rather than an open loop.

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