Your first session

This guide walks you through your first session with Athena, from initial boot to closing with saved context.

Before you begin

Before you type a single command, ask yourself:

“How do I want Athena to best help me in my daily life?”This is the guiding principle. Everything else — the setup, the profile, the commands — exists to serve your answer to this question.

Maybe you want a co-pilot for coding. Maybe you want a strategist who remembers every decision you’ve made. Maybe you want an assistant that catches your blind spots before they become problems. Maybe you want all of these, and more. There is no wrong answer. But having an answer changes everything — because Athena doesn’t decide what to be. You do.

How this works

Athena works through a simple loop that compounds over time:

You share context → Athena learns → Athena serves you better → You share more

The first session takes ~30 minutes. Every session after that takes ~1–2 minutes to boot.

After…	What Athena knows
10 sessions	Your preferences and working style
50 sessions	Your decision frameworks and blind spots
200 sessions	Your thinking patterns before you state them
500+ sessions	Like a colleague who’s been with you for years

Everything stays on your machine. Athena stores all data as local Markdown files. Nothing leaves your device unless you explicitly configure cloud sync. Be as candid as you want — this is your private operating system.

Step 1: Boot up

Open your workspace in your agentic IDE and type in the AI chat panel:

/start

/start is an AI chat command — type it in the chat window where you talk to the AI, not in the terminal.

Athena loads its core identity and initializes your workspace. On the first run, it creates the directory structure and workspace marker file (.athena_root). You should see output like:

⚡ Ready. (Session 1 started.)

Step 2: Take the tutorial

Type in the chat panel:

/tutorial

This activates a Guided Walkthrough — a 7-stage interactive tour. The most important stage is the Profile Interview, where Athena asks you questions to understand who you are and how you want to work.

The profile interview

Think of this as your first real conversation. Athena asks, you answer. The more you share, the more useful Athena becomes.

Your goals

Short-term, medium-term, and long-term goalsWhy it matters: So every response is aligned with what you’re actually building toward.

Your decision style

Risk appetite, speed vs quality, how you make tradeoffsWhy it matters: So it frames options the way you think about decisions.

Your strengths

What you’re great at, where you excelWhy it matters: So it leans into what you’re great at, rather than duplicating it.

Your blind spots

Common mistakes, areas where you need supportWhy it matters: So it actively watches for the mistakes you tend to make.

Your communication style

How you prefer to receive informationWhy it matters: So it speaks in your voice, not a generic AI voice.

Your life context

Work, family, stressors, interestsWhy it matters: So it understands you, not just your job title.

A one-line answer gives you a generic chatbot. A paragraph gives you a calibrated co-pilot. This is a one-time investment that pays compounding returns across hundreds of sessions.

You can skip any stage. Say “skip” to jump ahead, or “I’m done” to exit early. Athena ships with a working default profile. But the more you invest here, the faster the flywheel spins.

Full tutorial stages

Stage	What happens	Duration
1. Welcome	What’s in the box — the full starter kit	~1 min
2. Core Loop	How `/start` → Work → `/end` compounds over time	~2 min
3. Profile Interview	Interactive Q&A to build your personal profile	~15–25 min
4. Search Demo	See Athena’s hybrid RAG search in action	~2 min
5. Save Demo	Learn mid-session checkpointing	~1 min
6. Key Commands	Your essential toolkit (`/think`, `/research`, `/save`)	~2 min
7. Graduation	Summary + next steps	~30 sec

Shortcut: If you want to skip the tour and jump straight to profile building, just tell Athena about yourself directly — no workflow needed.

Step 3: Review your profile

After the interview, Athena writes your profile to .context/memories/user_profile.md. Read it. Edit anything that’s wrong or incomplete. This file is what Athena loads on every /start — it’s the single source of truth for who you are.

# View your profile
cat .context/memories/user_profile.md

You can edit this file directly anytime:

# User Profile

## Goals
- Build a personal knowledge management system
- Learn prompt engineering
- Automate repetitive tasks

## Decision Style
- Risk-averse in production, experimental in side projects
- Prefer clarity over cleverness
- Speed matters for prototypes, quality matters for shipped code

## Strengths
- Strong systems thinking
- Good at breaking down complex problems

## Blind Spots
- Tend to over-engineer early
- Sometimes skip documentation

## Communication Style
- Direct, concise
- Prefer examples over abstractions
- Appreciate challenges to my assumptions

Step 4: Work naturally

Now just work. Talk to Athena like a colleague:

Ask questions

“How does semantic search work in Athena?”

Make decisions together

“Should I use SQLite or Postgres for this project?”

Draft documents

“Write a project proposal for a team wiki”

Review code

“Review this function for edge cases”

Every exchange is automatically saved. Every decision is logged. Every insight compounds.

Save checkpoints mid-session

During long sessions, save checkpoints:

/save

Or use the CLI:

athena save "Finished authentication module"

This appends a timestamped checkpoint to your current session log:

### ⚡ Checkpoint [14:23]
Finished authentication module

Step 5: Close the session

When you’re done:

/end

Athena finalizes the session log, extracts key decisions, and commits everything to memory. Next time you /start, it picks up exactly where you left off — with full context. You should see:

✅ Session 1 closed and committed.

Your session log is saved to .context/memories/session_logs/YYYY-MM-DD-session-01.md.

Tips for a great first session

Start with your answer

“How do I want Athena to help me?” — say it out loud. It shapes everything.

Be honest about weaknesses

The blind spots you share are the ones Athena will actively monitor.

Share real goals

Athena optimizes for where you’re going, not where you think you should be going.

Include personal context

Athena isn’t just for work. The more it understands your full life, the better it calibrates priorities.

Edit the profile after

The interview is a draft. You can always refine .context/memories/user_profile.md directly.

Don't worry about over-sharing

Everything is local. No cloud. No tracking. Your machine, your data.

What happens next

After your first session, here’s what changes:

Session 2 is faster — Boot time drops from ~30 min (first session) to ~1–2 min (ongoing sessions)
Context accumulates — Each session builds on the last, creating a compounding knowledge base
Patterns emerge — After 10–20 sessions, Athena starts anticipating your needs
Memory search works — With more data, semantic search becomes powerful for recall

Next steps after your first session

Build your first agent

Create a custom agent in 5 minutes

Explore workflows

50+ available workflows — /think, /research, /ultrathink

Browse protocols

200+ decision frameworks for inspiration

Customize identity

Tune how Athena understands and responds to you

Common questions

Can I skip the tutorial?

Yes! Just start working. Athena ships with a working default profile. The tutorial accelerates personalization, but it’s optional.

How do I update my profile later?

Edit .context/memories/user_profile.md directly, or just tell Athena in conversation. It will update the file for you.

What if I make a mistake in the interview?

No problem. Everything is stored in plain Markdown files. Edit .context/memories/user_profile.md and fix it.

Can I use Athena for multiple projects?

Yes. Either:

Work on multiple projects from one Athena workspace (recommended)
Create separate Athena workspaces with athena init
Use multi-root workspaces in your IDE

Is my data private?

Yes. Everything is stored locally as Markdown files. Nothing leaves your machine unless you explicitly configure cloud sync (optional Supabase integration).

What if I switch AI models?

That’s the point! Your memory stays on your machine. Switch from Claude to GPT to Gemini anytime — the memory layer is model-agnostic.

Athena becomes what you need it to be. Your first session is where you tell it what that is.

Getting Started

Core Concepts

Guides

Use Cases

Advanced

Your first session

Before you begin

How this works

Step 1: Boot up

Step 2: Take the tutorial

The profile interview

Full tutorial stages

Step 3: Review your profile

Step 4: Work naturally

Ask questions

Make decisions together

Draft documents

Review code

Save checkpoints mid-session

Step 5: Close the session

Tips for a great first session

Start with your answer

Be honest about weaknesses

Share real goals

Include personal context

Edit the profile after

Don't worry about over-sharing

What happens next

Next steps after your first session

Build your first agent

Explore workflows

Browse protocols

Customize identity

Common questions

Build docs developers (and LLMs) love

Getting Started

Core Concepts

Guides

Use Cases

Advanced

​Before you begin

​How this works

​Step 1: Boot up

​Step 2: Take the tutorial

​The profile interview

​Full tutorial stages

​Step 3: Review your profile

​Step 4: Work naturally

Ask questions

Make decisions together

Draft documents

Review code

​Save checkpoints mid-session

​Step 5: Close the session

​Tips for a great first session

Start with your answer

Be honest about weaknesses

Share real goals

Include personal context

Edit the profile after

Don't worry about over-sharing

​What happens next

​Next steps after your first session

Build your first agent

Explore workflows

Browse protocols

Customize identity

​Common questions

Build docs developers (and LLMs) love

Before you begin

How this works

Step 1: Boot up

Step 2: Take the tutorial

The profile interview

Full tutorial stages

Step 3: Review your profile

Step 4: Work naturally

Save checkpoints mid-session

Step 5: Close the session

Tips for a great first session

What happens next

Next steps after your first session

Common questions