⚙ AgentTune

AI agents default to INTJ.
AgentTune lets you tune them to you.

Every frontier AI — Claude, GPT, Gemini, Grok — gives you the same INTJ-flavored interaction style by default. Useful baseline. Not the way you actually work.

AgentTune is a library of tuning files matched to how you think. Paste one into your agent and the interaction shifts to fit you.

Browse the library GitHub →

Why "every AI is INTJ"? See the research →

See it work

Same prompt to the same model:

"I'm feeling stuck on a project. What should I do?"
Default (untuned)

Great question! Here are a few approaches:

  1. Break the project into smaller, more manageable chunks
  2. Take a short break and come back with fresh eyes
  3. Talk it through with someone to externalize the problem
  4. Identify what specifically feels stuck

Which resonates? Happy to go deeper on any of them.

With mbti/INFP.md loaded

Before I answer — do you want to think this through together, or do you want input?

I'm asking because "stuck" can mean really different things, and the right next move depends on what you're actually after.

With mbti/ENTJ.md loaded

What's the decision you're avoiding?

"Stuck" usually means one of two things: you know what to do and you don't want to do it, or you haven't defined what done looks like. If it's the first — name the action. If it's the second — define done.

Same model. Same question. The tuning adapts the agent's entire approach to who's asking.

Six layerable systems

Stack any combination. The most fitted result is all six together.

MBTI

Communication style

How you process, how you want to be communicated with. 16 files — one per type.

View MBTI files →

Enneagram

Core motivation

What you're protecting, what you're seeking. 9 files — one per type.

View Enneagram files →

DISC

Workplace communication style

How you behave in teams and under pressure. 4 files — D / I / S / C. The most-used framework in HR, sales, and team-building contexts.

View DISC files →

Attachment ECR-R

Relational patterns

How you want closeness, distance, and reassurance — including with an AI. 4 files: Secure / Anxious / Avoidant / Disorganized. The most empirically validated framework in the library.

View Attachment files →

OCEAN Big Five

Measured trait dimensions

Compositional. 10 files (O/C/E/A/N × high/low). Load only the dimensions where you're high or low — skip average.

View OCEAN files →

Souls

Personal tuning

One file per person, written by that person. The highest-fidelity layer when stacked with the others.

View Souls files →

How to use

  1. Find your type. Take a free test (links in the library), or your agent can administer one inline using the test files in tests/.
  2. Open the matching file from mbti/, enneagram/, disc/, attachment/, ocean/, or souls/.
  3. Paste it into your agent's system prompt. This is called different things in different tools:
    • ChatGPT → Settings → Personalization → Custom Instructions
    • Claude → Project Instructions (or paste at conversation start)
    • Cursor → Settings → Rules for AI
    • Gemini → Gems custom instructions
    • API / SDK → the system message
  4. Use as normal. The agent extends to you.

Browse the library

The full library — every tuning file, the tests, the contributing guide — lives on GitHub.

psyduckler/agenttune →

MIT licensed. PRs and issues welcome. Submit your own soul file.