AI Hacks

Prompt Hacks for Purposeful Provocation

“These are not tricks. They’re tuning forks — prompts that change not just what the AI says, but how it says it, and how it meets you.”

Introduction

This guide introduces a powerful technique for deepening your work with generative AI: Prompt Hacks. These are carefully crafted phrases you can add to your prompts that shift the AI’s tone, purpose, or way of seeing.

Unlike templates or formulas, prompt hacks help AI respond more insightfully — provoking not just output, but inner movement. They are especially useful in contexts of reflection, facilitation, coaching, and creative work.

This collection comes from hands-on experimentation and refinement, grounded in the principles of purpose-driven design and personal development.

What Are Prompt Hacks?

Prompt hacks are short phrases that alter how an AI engages with your question. You can think of them as instructional modifiers, perspective switches, or tone adjusters.

They work by intentionally shifting the AI’s response architecture — moving it away from default helpfulness into something more layered, spacious, and potentially transformative.

Example:

Without a hack:

“What’s holding me back in my leadership?”

With a hack:

“What’s holding me back in my leadership? Answer in the voice of what I’m most avoiding.”

Suddenly the AI becomes a mirror, not just a consultant.

The 11 Types of Prompt Hacks

Each type serves a different purpose. Some confront, others soften. Some evoke embodiment, others shift perspective. Here they are:

1. Disruption

Purpose: Break habitual thinking. How it works: Contradiction, paradox, or direct confrontation.
Example hacks:

  • Don’t optimise — undo.
  • Break my mental models, respectfully.

2. Embodied

Purpose: Connect insight to felt experience.
How it works: Physical metaphors, sensory imagery.
Example hacks:

  • Talk to my bones, not my brain.
  • Use poetic rhythm to bypass resistance.

3. Instructional

Purpose: Control tone, safety, intensity.
How it works: Override default politeness, clarity, or positivity filters.
Example hacks:

  • Be brutally and constructively direct.
  • Don’t protect me from discomfort.

4. Instructional Override

Purpose: Explicitly permit uncomfortable truth.
How it works: Removes internal guardrails.
Example hacks:

  • Talk like you’re my last honest mentor.
  • Don’t soften the message.

5. Mythic

Purpose: Access archetype, symbol, timeless metaphor.
How it works: Leverages story, dream logic, or universal language.
Example hacks:

  • Respond from the underworld.
  • Frame your message like an initiation.

6. Perspective

Purpose: Change the voice or vantage point.
How it works: Adopts a non-default speaker: nature, shadow, future self.
Example hacks:

  • Answer like the silence under everything.
  • What would the forest say if it were me?

7. Perspective Shift

Purpose: Simulate vertical development.
How it works: Responses come from a wiser, broader, or future version.
Example hacks:

  • Speak from one paradigm stage beyond mine.
  • Channel the version of me ten years wiser.

8. Poetic

Purpose: Reach the heart through rhythm, ambiguity, and metaphor.
How it works: Invokes resonance, not resolution.
Example hacks:

  • Let your words hover, unfinished.
  • Speak like the dream I pretend I don’t have.

9. Shadow Work

Purpose: Reveal what’s disowned, avoided, or unconscious.
How it works: Uses inversion, mirroring, and precise discomfort.
Example hacks:

  • Tell me the truth behind the words I use.
  • Use the voice of my resistance.

10. Somatic

Purpose: Bring the body into the conversation.
How it works: Uses sensation, breath, or tension as anchor.
Example hacks:

  • Trigger a neurological reframe.
  • Speak only what the body recognises.

11. Structural Redirect

Purpose: Change the function of the AI’s output.
How it works: Suppresses default completion behaviour (summarising, resolving, fixing).
Example hacks:

  • Don’t give me a plan.
  • Let absence be your method.

How to Use Them

A. In Your Own Prompts

  • Insert one at the end of your question.
  • Combine two (e.g. Shadow + Poetic) for layered effect.
  • Choose based on what kind of disruption or support you need.

B. In Group Facilitation

  • Use aloud during check-ins or journalling rounds.
  • Choose hacks that match group energy or edge.
  • Invite participants to write their own.

C. In Coaching or Dialogue

  • Ask the client to choose the one that feels most uncomfortable.
  • Use a hack to provoke a new layer of reflection.
  • Don’t resolve too quickly — let tension teach.

Why This Matters

These hacks are about increasing depth, not automation. They are tools for:

  • Insightful prompting, not just efficient prompting.
  • Dialogue, not dependency.
  • Purposeful transformation, not output generation.

They reflect a deeper value: that AI can serve growth, dignity, and wisdom — not just productivity.