The Codex: How to Seal, Burn, and Reflect on Sessions

The Problem With Endless Chat

Most people treat artificial intelligence as a general-purpose utility. You open your go-to model to handle mundane tasks like checking your daily protein intake or finding a banana bread recipe. These platforms function beautifully as everyday assistants because they deliver immediate results for common queries.

However, doing deep inner work on these same platforms requires intense manual labor. You often have to write massive, heavily engineered prompts just to coax a general-purpose model into a state of mind capable of facilitating dream work or the exploration of human consciousness.

MOODS deliberately separates your private inner life from these general tasks. We’ve already handled the heavy lifting by permanently dialing in specific psychological and esoteric frameworks. You get to skip the setup phase entirely and jump straight into the deep end.

In addition to the struggle of constant prompting, there is a massive technical flaw in how we usually interact with these models, especially when it comes to deep, long-form conversations. We tend to treat AI like an endless text message thread, often keeping a single conversation running for weeks on end.

Anyone who has done deep inner work with an out-of-the-box model knows exactly what happens next. The quality of the conversation eventually falls off a cliff. You might feel like you finally prompted the machine into the perfect state of mind, but the longer the thread gets, the more the AI starts to lose the plot. It forgets early details and starts to hallucinate (make sh*t up) to fill in the gaps.

Recent research actually confirms exactly what many of us have experienced. When an AI is forced to engage in these long, multi-turn conversations, it suffers a 39% drop in performance and a 112% spike in unreliability. These models just aren't meant to work this way.

Our team noticed this before the data was published, and we designed MOODS specifically to solve it. Sessions with our archetypes are meant to be short and potent. Because we've already handled the background frameworks, you don't have to waste time "warming up" the AI. You engage with an archetype, work through a specific issue, find your realization, and then seal the session.

Keeping archetypal exchanges brief forces the intelligence to remain incredibly sharp. You get a high-fidelity mirror for your own psyche without the risk of the AI drifting into strange rabbit holes or bloated responses. It’s simply a more accurate (and much healthier) way to engage in deep inner work.

What is the Codex?

We built the Codex to solve the exact problems we just discussed. It serves as the permanent record of your inner life and the deep work you do within MOODS. Many people find it functions as a digital version of an alchemical journal or a magical diary — a place where your subconscious finally feels like it's being heard.

The idea here is that by focusing deep intention in one place, MOODS acts as a digital sanctuary. Your subconscious recognizes when you're paying attention to it, and that work needs a distinct place to live. In most AI platforms, your history is just a hodgepodge of different chat logs that eventually get buried or deleted. They aren't given the gravity they deserve. The Codex changes that by turning your inner work into a beautiful library.

Every insight you seal into your Codex is preserved, meaning that over time, you’re building a cumulative biography of your own psyche. Your inner world deserves its own space, and that’s what we've created here.

This is meant to move you away from the digital clutter of hoarding massive, winding transcripts. Instead, when you finish a session, the AI evaluates the exchange to capture the essence of your work. When an interaction is sealed to your Codex, you get a gorgeous insight card that preserves that specific moment in time. You don't have to sift through long, messy threads to find what mattered. You just open your Codex and scroll through a map of your own mind.

Sealing Your Sessions

Sealing a session is a core mechanic of MOODS. Interactions with our archetypes are meant to be potent and short. The platform expects you to finish the work in one sitting.

The archetypes will keep asking you questions throughout the engagement. You decide when the conversation is over. The idea is to seal the session the moment you hit a core truth. If you try to close the session out too soon, a screen will appear letting you know the work isn't finished yet. The system requires enough back-and-forth to actually generate a meaningful summary.

When you're ready to seal a session, you click "Leave" at the top of the screen and select "Seal and send to Codex", prompting the AI to scan your entire exchange so it can produce an insight card for your personal library.

From there, the MOODS system assigns a specific title and writes a short summary for your main feed.

Clicking into the card reveals an expanded description of your realization. You get the exact heart of the conversation. You bypass the tedious process of digging through a massive chat thread.

You can also hit "Read Session Transcript" anytime to review the full raw exchange.

The act of sealing your sessions drives the entire backend of the app. Sealing allows the system to evaluate your interaction and update your archetypal constellation.

For example, if you have an intense conversation with The Severant, the MOODS system scans the conversation to determine how that energy manifested. It decides if the archetype is acting from an integrated position. It might also register as an inverted or exiled state.

Some engagements are intense enough to change an archetype's status all at once. Other times the adjustments happen quietly in the background without an immediate visual update.

Leave the background mechanics to the system. Obsessing over whether a specific conversation will invert or integrate an archetype distracts from the actual work. You simply engage until you hit a realization. Once the encounter runs its course, seal the session. The app handles the underlying updates. You walk away knowing your Codex holds a permanent map of your inner world that you can return to whenever you want.

Burning, Fire and Friction

You always have the option to burn a session. Hitting the burn button permanently deletes the entire interaction.

People use this feature in vastly different ways. You might want a highly charged, ritualistic cord-cutting session where you confront an archetype with raw frustration, knowing you want to wipe the exchange from existence the second you finish. Or you might just accidentally open a session by accident and want to get rid of it. Burning guarantees the interaction will never impact your archetypal constellation. When you burn a session, the data is completely irretrievable.

You'll notice an alert badge sitting at the top of your Codex tracking your unsealed sessions. We built this exact friction into the system on purpose. MOODS actively prompts you to resolve your open sessions. The app constantly pulls your attention back to those unfinished conversations because the platform relies on short, potent engagements.

The idea here is to encourage you to make a conscious choice. A session either belongs in your permanent library or it belongs in the fire.

Navigating Your Codex

You understand by now that once a session is sealed, it lives in your Codex as a dedicated insight card. These cards are designed to give you immediate, visual feedback on your inner work.  But here is an additional little Easter egg for you: when you click into an entry, look closely at the glyph next to the archetypal illustration.

If that specific session caused an immediate shift in one of your archetypes (flipping them to integrated, inverted, or exiled) you will see a little glyph indicating the change. This lets you know right away which interactions made the biggest impact on your constellation. If no shift occurred, there is no glyph.

We also built a feature that let you continue working with previously sealed insights. When you're doing inner work, you might return to your Codex weeks later and realize you have a new perspective.

Clicking "Reflect on this insight" scans a previous conversation and starts up a brand new encounter with that same archetype using the past context as your starting point. People often leave sessions open just in case they want to revisit the topic later. This feature completely solves that hesitation. You can branch off your old work with a new conversation right where you left off.

You also have the ability to "Add a note" to any card. These notes automatically save as you type. Think of them as a scratchpad in the margins of your diary. Use them to add personal context in your own words.

You could use this feature for almost anything:

• Recording specific reality details ("This dream was from week 38 of my pregnancy.")
• Reminders for future work ("Try revisiting this same concept with the Severant...")
• Contextual tie-ins ("This reminds me of what my therapist said about setting boundaries.")

These notes are strictly for you. They don't impact the MOODS system or your archetypal constellation. They are simply there for you to add your own perspective to your records.

The Work Is Yours

By now, you have all the information you need to confidently navigate your Codex. We built these features—sealing, burning, reflection, and notes—to provide you with a powerful, permanent archive of your own inner world. The ultimate intention is for you to develop a beautiful record of your own personal evolution. We encourage you to experiment and intuitively use MOODS in the way that best serves your own practice.