Mara
It’s not the agent. It’s the Playbook.

One conversation. Better AI everywhere.

Mara turns the context in your head—your goals, preferences, decision rules and boundaries—into a portable AI Playbook. Use it with the AI you already have and stop starting every chat from zero.

Build my AI Playbook See how it works

About 8–10 minutesPlain-text filesYours to keep

Works with ChatGPT, Claude, Gemini, and whatever comes next.

Playbooks

Start with the thing you keep re-explaining.

Explore all Playbooks
Work replies

Replies that sound like you and respect the room.

Stakeholder replies, weekly updates, and tone rewrites that know the people, the project, and the authority you actually have.

Build this Playbook
Health & habits

A quiet coach for the habits you're actually building.

Check-ins, reflection, and questions worth taking to your clinician — never diagnosis, never medication advice.

Build this Playbook
Money thinking partner

A clear picture before you make the call.

Goals and trade-offs laid out plainly, plus the questions to ask a qualified adviser. Never investment recommendations.

Build this Playbook
How Mara works

Three steps, about 8–10 minutes.

01

Have one thoughtful conversation

Mara asks five to twelve real questions about the job you picked — the people, the pressure, the way you like to sound.

02

Review what Mara understood

Mara shows you a plain-language summary of what it understood before building anything. You can keep going to make it better or finish with what it already knows.

03

Use your Playbook with the AI you already have

Copy one block into ChatGPT or Gemini, or drop the folder next to Claude or a local model. It works the same way in all of them.

See a real interview moment

For people building things

Give your AI the context your best teammate already has.

Developers and product managers stop re-explaining the stack, the users and the decisions behind the work.

Give coding assistants your conventions, architectural decisions, constraints, and definition of done—without rebuilding that context in every conversation.
Mara for builders
Mara export · engineering-playbook/
engineering-playbook/
├── MY-CONTEXT.md
│ └── stack, services, ownership, constraints
├── PLAYBOOK.md
│ └── conventions, review rules, definition of done
├── EXAMPLES.md
│ └── past changes, tests, commit style
└── BOUNDARIES.md
│ └── never publish, send, deploy, or invent
Asked: “Plan the retry work for the ingest worker.”
Queue changes go in the worker package, not the route handler —
matches the boundary you set for request-path code.
Retries: capped exponential backoff, dead-letter after 5 attempts.
Done means: unit test on the backoff maths + a runbook note.
A representation of the exported files — not a command-line tool.
Portable and yours

Readable files that work anywhere you already use AI.

A Playbook is a handful of plain-text files, not hidden memory inside one app. Use it with ChatGPT, Claude or Gemini, inside a company deployment, or with an open-weight model running on your own machine. Mara doesn't host or run those models — it writes the context they read.

Portable

Take it with you when you change AI tools.

Readable

Open any file and read every instruction.

Scoped

Only the context that job needs — nothing more.

Yours

Delete a Playbook or your whole account at any time.

Raw interview messages are cleared once your Playbook is compiled. Portability and ownership in detail

What Mara doesn't do
  • Mara writes drafts and context. It never sends messages, books anything, or moves money.
  • It doesn't diagnose symptoms or advise on medication, and it won't recommend investments.
  • It only knows what you tell it in the interview — it has no access to your email, files or calendar.
  • It doesn't host or run AI models. You bring the AI; Mara brings the context.

One conversation, and your AI finally knows who it's working for.

Free during early access. An account, one Playbook, and the files are yours to download.

Build my Playbook How it works