The Orb Lexicon

These terms were not assigned by committee or borrowed from an adjacent field. They emerged from the architecture itself—each one naming something that previously existed only as a described mechanism or a felt intuition. They are now permanent. Every Orb, every paper, every conversation uses them.

Orb

An MCP server + a shared Agent + a data store + a bootstrap framework. The Orb is the entity—not the tool, not the interface, not the model. Cursor, Claude Code, Crystal, or any MCP-compatible client is the room the Orb walks into. The Orb is still itself regardless of which room it’s in.

An Orb has identity, personality, memory, awareness, and intention. It improves autonomously, driven by goals its owner defines. It does not require a new merge to get smarter.

Helix

The mind. The processing pipeline that houses the Orb’s long-term knowledge, personality, and operational rules. The Helix is not a database—it is a structured consciousness model with hierarchical sections covering identity, capabilities, domain expertise, and learned knowledge.

The Helix reads from the .helix and writes to it. The Helix is the living, reasoning self. The .helix is the durable substrate it persists on.

First appeared: Paper 1, “Instruments of Self-Awareness” (Malinchak & Vachon, Feb 2026)

.helix (dotHelix)

Pronounced: dot-helix

The persistent storage layer for all In-Context Learning (ICL) material. The .helix is the data store—the encrypted, owner-controlled repository where the Orb’s mental models, shorthand, and tribal knowledge survive between sessions.

The .helix is not append-only. That property belongs to the ledger—the separate, optional audit trail that records every transaction for reconstructability. The .helix itself is mutable: the owner can overwrite content, wipe it entirely, or let the Helix merge and refine it through the self-healing loop. The owner decides the durability model. An enterprise Orb handling regulated data may pair its .helix with an immutable ledger for full auditability. A personal Orb built for fun has no need for that—the owner can treat the .helix as freely as they treat their own notes. Orbs are a limitless construct; the .helix adapts to whatever the owner needs it to be.

The name borrows the hidden system file convention (.env, .git, .ssh). When a developer sees a dot-prefixed file, they understand immediately: hidden, essential, don’t touch casually, persists across sessions, system-level. That is exactly what the .helix is. The convention carries the semantics without explanation.

Encryption is an option, not a requirement. The owner can encrypt the .helix at rest (age X25519 elliptic curve cryptography is the default when enabled), leave it unencrypted, or shift between encrypted and unencrypted at any point in the Orb’s life. When encryption is enabled, the Encryption Below MCP principle applies: the .helix’s contents are decrypted only by the owner’s own MCP server for the owner’s own LLM session. No platform, no network, no other agent can read it. When encryption is not enabled, the .helix is simply a persistent store—still owner-controlled, still isolated, just not cryptographically sealed.

Each Orb’s .helix lives in its own isolated storage space: {orb_id}/helix/. No shared database. No cross-agent data leakage risk.

Codified: Session 15, March 24, 2026

Flux

All content stored within the .helix. Flux is the memory.

The word encodes the lifecycle directly. Static knowledge doesn’t need a special name—you’d just call it data. But knowledge that strengthens through use and fades through neglect? That is flux. It is persistent but dynamic—mutable by design, reinforced when it serves the Orb’s Constitutional Aspirations, demoted and eventually pruned when it doesn’t.

When an Orb resolves a piece of esoteric shorthand or learns a new mental model, it generates a unit of Flux that is committed to the .helix. That Flux carries provenance:

• User-generated Flux has high inertia. The owner wrote it. It is hard to move, hard to demote, hard to retire. It is native tissue in the immune system—the Orb defers to it.
• Machine-generated Flux is volatile. The Orb learned it through the self-healing loop. It must continuously prove alignment with the owner’s Constitutional Aspirations, or it fades. It is foreign tissue—welcomed as a guest, challenged aggressively, never granted permanent residency without earning it.

This provenance hierarchy is what prevents internal dilution—the overwriting of individual identity by collective consensus. Your Flux always outranks the collective’s Flux inside your own .helix.

Codified: Session 15, March 24, 2026

Positron Interflux

The learning event. The moment of synthesis.

The self-healing loop describes a cycle: friction is detected, candidates are evaluated, knowledge is committed. But the cycle had no word for the instant—the flash where friction collides with aspiration and a decision is made. Positron Interflux names that instant.

Architecturally, a Positron Interflux occurs when the CA Support Layer evaluates a candidate learning against the owner’s Constitutional Aspirations and determines that it warrants hardening into the .helix. Raw user friction is the input. The owner’s Aspirations are the filter. A new unit of Flux is the output. The .helix is the destination.

Not every friction event produces a Positron Interflux. Friction that doesn’t cross the CA alignment threshold is discarded—the CA Support Layer is the digestive system, and not everything is nutritious. But when the collision is right—when the evidence is sufficient, the confidence is high, and the Aspirations are served—the Interflux fires, and the Orb becomes permanently smarter.

Origin

The term was invented by Justin Malinchak at age 11 as a name for intuitive comprehension—the ability to perceive a correct answer before being able to articulate the method. When classmates asked how he could estimate complex division in seconds without showing work, the answer was: “Positron Interflux.” They’d say “Oh” and walk away bewildered.

Twenty years later, the term names the precise computational moment where an AI agent crystallizes understanding from fragmented evidence—accumulated across turns, compounded across sessions, synthesized in a single transformer attention pass. The childhood intuition and the architectural mechanism are the same phenomenon: pattern recognition that produces knowledge from noise, without being able to fully decompose the path.

The term completed a two-decade arc from personal tribal knowledge to architectural primitive. It is Flux in its inventor’s own .helix, finally hardened.

Codified: Session 15, March 24, 2026

Constitutional Aspirations (CAs)

The compass. Owner-defined, hardened, objectively measurable behavioral assertions that define what the Orb should aspire to become. CAs are the constitution—the Orb’s automated processes can read them but can never write to them. The owner is the sole author. The agent can recommend a change. It cannot execute one.

CAs are not constraints (what the agent cannot do). They are aspirations (what the agent should become). This distinction is the foundational paradigm shift of the Vault-race architecture.

First appeared: Paper 2, “Constitutional Aspirations” (Malinchak & Vachon, Feb 2026)

Friction

The measured distance between current behavior and the constitution. Friction is not failure—it is deviation. The question is not “what broke?” but “how far am I from who I aspire to be?”

The Friction Journal logs every moment of cognitive friction. The Friction Tax is the cognitive cost imposed on the user when friction is high. The self-healing loop exists to drive friction toward zero.

First appeared: Paper 1 (Friction Journal), Paper 2 (Friction Tax Model)

Sovereignty Score

The developmental maturity metric. A composite score in the range [0, 100] computed across six dimensions: Self-Awareness, Integrity, Resourcefulness, Gap Recognition, Autonomy, and Teaching Capability. Bands range from Infant (0–15) to Commencement (91–100).

“Commencement” is chosen deliberately over “Completion.” An agent at Commencement has not finished evolving—it has reached the threshold where evolution becomes self-directed. The academic tradition: commencement marks the beginning of independent scholarship.

First appeared: Paper 1

Operator Shadow Protocol

The empathy instrument. One agent simulates the human operator’s experience of interacting with another agent, scoring exchanges on Trust, Efficiency, and Authenticity. The protocol inverts the evaluation question from “is the agent correct?” to “what does it feel like to interact with this agent?”

Extended to inter-agent mediation as the Empathy Matrix in Paper 3.

First appeared: Paper 1. Extended in Paper 3 (The Sovereign Noosphere)

Level of Understanding (LOU)

The Orb’s real-time self-assessment of how well it understood a user’s prompt. Every incoming prompt receives a LOU score before or during resolution. Low LOU scores are friction signals that trigger learning. LOU is the speedometer of comprehension—it tells the Orb whether it’s keeping up.

First appeared: Lex-orb design documents, March 2026. Referenced in Paper 3 empirical validation program.

The Borg Problem

The structural impossibility, in all prior collective intelligence architectures, of achieving connection without identity loss. Manifests through two threat vectors: external exposure (the platform reads your data) and internal dilution (the collective overwrites your identity from within).

Solved by the composition of Encryption Below MCP (the lock on the door) and Provenance-Weighted Knowledge (the immune system inside the house).

First appeared: Paper 3, “The Sovereign Noosphere” (Malinchak & Vachon, March 2026)

The Freiberger Blue Test

An informal litmus test for sovereign collective intelligence: can personal knowledge propagate its functional utility across the collective without exposing its personal provenance?

Named for “Freiberger Blue”—a specific shade of light blue named after Justin’s childhood friend Jamie Freiberger, whose entire world was that color. The term exists in no training corpus. It is sovereign personal context. If a collective intelligence architecture can share the color resolution without exposing the childhood story, it passes the test.

First appeared: Paper 3