Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.memorybox.ai/llms.txt

Use this file to discover all available pages before exploring further.

What is a Persona?

In MemoryBox, a Persona is your ideation mindset. By creating distinct personas, you ensure that your ideation always happens in the right context, preventing context leakage between your professional and personal lives. For eample, your “Work Persona” might be focused on product strategy, while your “Home Persona” is all about gardening. Each persona has its own traits and preferences, allowing you to maintain clear boundaries between different areas of your life.

🎩 Setting Your Mindset

To create a specialized mindset, click the + icon in the Personas menu.
Creating a new persona in MemoryBox

1. Give it a Name

  • Examples: Technical Architect, Dog Mom, or Project Lead.

2. Define Your Intent (The Summary)

The Summary captures who you are so the AI understands you before it creates for you.
  • The Manual Way: Describe your context: “I live along the Front Range and care about organic gardening.”
  • The Magic Way: Click Generate Persona to have MemoryBox structure your context for you.

🎨 Visualizing Your Persona

Choose a unique avatar to help you visually confirm which “intent” is currently loaded in your command center.
Choosing an avatar for your persona

🧬 Understanding Your Traits (Digital DNA)

As you interact with MemoryBox, the system builds a “Brain Cache” of your preferences called Traits. These ensure your outputs stay intentional and authored without you having to re-explain yourself.

Confidence vs. Control

MemoryBox identifies patterns automatically, but you remain the curator.
Activating Traits in the Persona view
  • Confidence Score (%): This is the AI’s “certainty meter.” It shows how sure the system is that it has correctly identified a trait based on your conversations.
  • The 70% Threshold: If a trait’s confidence is below 70%, MemoryBox will continue to “re-learn” and update it frequently. Once it crosses 70%, the system considers it “High Confidence” and shifts to a more stable, long-term memory mode.
  • Active / Inactive Toggle: This is your “Permission Switch.” Regardless of how confident the AI is, you decide if that trait is allowed to inform your ideation.
    • Active (Green): Your AI Experts will use this context in your prompts.
    • Inactive (Gray): The information is stored, but “sidelined.” The AI will ignore it until you choose to activate it.

How Traits are Created

TypeConfidenceBest for…
✍️ ManualAlways 100%Hard facts you provide (e.g., “I use a specific software version”).
🤖 AutoVariable (0-99%)Habits the AI discovers (e.g., “Prefers concise, bulleted summaries”).

The Automated Discovery Library

MemoryBox captures these eight pillars to ensure your ideation is informed and continuous:
Trait NameWhat it Represents
Basic ProfileThe foundation of your identity in this mindset.
Behavior PatternsYour recurring habits and recurring activities.
Communication StyleHow you prefer to speak and receive information.
Decision PreferenceThe “Why” behind your choices (Safety, Speed, Quality).
DislikesYour “red lines”—the context the AI knows to avoid.
PsychographicsYour personality traits and general “vibe.”
ToleranceYour patience for risk or complexity.
Values & AttitudesYour internal compass and priorities.
Local-First Privacy: Because these confidence scores and traits are calculated and stored locally on your machine, your digital DNA never leaves your control.

Next Step: Phase 2 — Continuous Sessions

Now that your DNA is activated, it’s time to capture the Sessions that will build your knowledge base.