# Habit 2: Begin with the End in Mind → Frame the Output’s Role

#### What It Means in LLM Context

Covey’s original Habit 2 is about **vision before execution.**\
In LLM usage, it means:

Don’t prompt until you know what **you’ll do with the output.**

Are you writing for publication?\
Is this *scaffolding* for a bigger plan?\
Will you reuse it in a slide deck, reflection journal, or pitch?

A prompt without an output role is like building a bridge to nowhere.

***

#### The Hidden Power of Declaring Purpose

By stating your intended **use case**, you unlock:

* Better tone
* Better structure
* Better focus
* Better fidelity to your values or context

And you help the model **collapse ambiguity** into the right genre.

***

#### Common Unframed Patterns

| Prompt                    | Unclear End Use                            | Typical Problem              |
| ------------------------- | ------------------------------------------ | ---------------------------- |
| “Explain neural networks” | For kids? For peers? For hiring manager?   | Generic, mismatched tone     |
| “Write a bio”             | Twitter bio? Academic CV? Press kit?       | Wrong length, wrong register |
| “Fix this code”           | For debugging? Refactor? Interview answer? | Missing constraints          |
| “Give me a metaphor”      | For teaching? For branding? For therapy?   | Style mismatch               |

***

#### Framing Tactics

| Tactic                      | Prompt Addition                                                                               |
| --------------------------- | --------------------------------------------------------------------------------------------- |
| **Declare purpose**         | “This is for a cultural studies course on digital identity.”                                  |
| **Declare ambiguity**       | “I’m curious about how avatars shape self-perception, but I don’t yet have a defined thesis.” |
| **Describe help needed**    | “I need help surfacing angles, assumptions, or research framings, not just summaries.”        |
| **Tag tone of exploration** | “This is early-stage scaffolding, not something I need to cite yet.”                          |
| **Set constraints**         | “Avoid overly technical papers or sources behind paywalls.”                                   |

***

#### Example Transformations

**Unframed Prompt:**

“Help me explain LLMs.”

**Framed Prompt:**

“I’m building a 90-second explanation for a non-technical CEO who fears LLMs are hype. I need a calm, concrete framing with 1 business use case, no jargon.”

***

**Unframed Prompt:**

“Give me pros and cons of going to grad school.”

**Framed Prompt:**

“I’m sketching a reflective guide for students unsure whether to pursue graduate school. I don’t want a generic list. Help me map the mental models people carry into this decision: identity signaling, debt-for-certainty tradeoffs, social delay vs. career acceleration. I want to expose what’s often unstated behind the ‘pro/con’ lists.”

***

**Unframed Prompt:**

“Design a settings menu for this app.”

**Framed Prompt:**

“This app helps neurodiverse users focus during study. The settings menu should feel calming and non-intrusive, with minimal text. I want layout ideas that reduce friction and reduce decision anxiety.”

***

#### Self-Check

* Do I know how I will **use** this output?
* Have I told the model who the **audience** is?
* Am I clear on the **tone and format** that will best serve my need?

If not, the model is guessing. And so are you.

***

#### Framing Is *Before* Prompting

This habit isn’t about rephrasing the prompt. It’s about doing **your own internal reflection** first.

Covey’s principle: “All things are created twice — first in the mind, then in the world.”\
LLM corollary: “All good prompts are created twice — first in your intent, then in the syntax.”

***

#### Mini-Taxonomy of Output Roles

| Role            | Framing Cue                         | Prompt Addition                                                       |
| --------------- | ----------------------------------- | --------------------------------------------------------------------- |
| **Draft**       | “I’ll revise this later.”           | “Give me a rough first pass with modular structure.”                  |
| **Public**      | “This is for sharing.”              | “Polish for external audience. Prioritize clarity over detail.”       |
| **Exploration** | “I don’t know what I think yet.”    | “Help me map the landscape of this issue without forcing a position.” |
| **Dialogue**    | “This will feed a conversation.”    | “Add openings for questions or reflective pauses.”                    |
| **Teaching**    | “This is for others to learn from.” | “Include a core principle, 1 analogy, and an actionable takeaway.”    |

***

#### Integration Cue

*Before every prompt, ask: “What role will this output serve in my broader work or life?”*

Framing isn’t decoration. It’s what separates **surface prompting** from **strategic cognition**.

***

#### Metaphor

The clearer your intent, the more the model can shape toward it.\
Purpose is the anchor that prevents drift.

***


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://praxistutor.gitbook.io/llm/reframing-the-7-habits-for-reflective-llm-use/section-2-the-seven-habits-of-reflective-llm-use/habit-2-begin-with-the-end-in-mind-frame-the-outputs-role.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.