Aura Hive — Visual Reasoning Index

This section provides a visual overview of Aura Hive from an advisory and reasoning perspective.

The diagrams in this section illustrate how observability signals, events, and interpretations relate conceptually.


Hive-Level Cellular Map

Abstraction Level: Level 2 (Cellular) — Shows ATCG-M nucleotide interactions within a Bee Advisor

spinner

Signal Metabolism Flow:

  1. Inbound Membrane (M) validates observability signals, detects prompt injection

  2. Aggregator (A) collects git diffs, Prometheus metrics, filesystem state

  3. Transformer (T) reflects on architectural purity via LLM reasoning

  4. Outbound Membrane (M) enforces business rules, prevents data leakage

  5. Connector (C) acts through Jules proteins (GitHub comments, NATS events)

  6. Generator (G) pulses chronicles to HIVE_STATE.md and event streams


How to Read These Diagrams

  • Diagrams describe interpretation, not execution

  • Solid arrows indicate signal metabolism (active processing through ATCG-M)

  • Dashed arrows indicate observation (humans reading state)

  • Subgraphs show cellular boundaries (complete ATCG-M organisms)

  • No diagram implies automation or authority

These diagrams answer the question:

"How does the hive understand itself through its fractal ATCG-M metabolism?"


🧬 Hive Metaphor Key

Aura Hive uses Technical Folklore — bio-organic metaphors layered over precise technical architecture. This table bridges newcomers (who need literal terms) and established contributors (who think in metaphors):

Bio-Organic Term
Technical Equivalent
Rationale
Code Reference

Bee

Autonomous Service

Self-contained, cooperative agents with agency

agents/bee-keeper/

ATCG-M

Service Architecture Pattern

DNA analogy for fractal 5-nucleotide structure

FOUNDATION.md:27

A (Aggregator)

Input/Sensor Layer

Collects signals from environment

dna.py:153-156

T (Transformer)

Reasoning/LLM Layer

Processes signals via Ona (DSPy, LangChain)

dna.py:160-163

C (Connector)

Output/Action Layer

Acts through Jules (Proteins/Skills)

dna.py:167-170

G (Generator)

Event/Chronicle Layer

Pulses state updates and events

dna.py:174-177

M (Membrane)

Input/Output Guards

Immune system: validates inputs, enforces rules

core/src/hive/membrane.py

Bloodstream

NATS Event Bus

Circulatory system distributes signals

FOUNDATION.md:14

Proteins

Skill Adapters (SkillProtocol)

Enzymes enable specific external interactions

dna.py:184-202

Sacred Chambers

Directory Roles

Poetic names for filesystem organization

dna.py:50-75

Genome

aura-core package

Immutable DNA (Protocols, BaseModels, Types)

packages/aura-core/

Nucleus

core

The sovereign brain

core/

Organs

components/proteins

Specialized skills (Solana, Telegram, GitHub)

components/proteins/

Citizens

agents/ + adapters/

Active Subjects (goals) + Passive Servants (translation)

agents/, adapters/

Ona

LLM Reasoning Engine

The thinking substrate (DSPy, LangChain)

N/A (conceptual)

Jules

External Action Engine

The doing substrate (API calls, I/O)

N/A (conceptual)

Why Metaphors?

  • Memorability: "Membrane guards against injection" is stickier than "input validation layer"

  • Consistency: Same pattern (ATCG-M) repeats fractally across all services

  • Aesthetic: Cyber-organic folklore creates a distinct architectural identity


Relation to Canonical Architecture

Visual diagrams are aligned with the canonical architecture defined in:

  • docs/FOUNDATION.md — Ontological hierarchy and ATCG-M metabolism

  • packages/aura-core/src/aura_core/dna.py — Protocol definitions

  • HIVE_STATE.md — Operational status and audit log

  • Reference implementations: agents/bee-keeper/, core/src/hive/


🧬 Hive-Level Maps (visual/hive/)

🔬 Component-Level Views (visual/components/)

🩸 Pipeline & Event Flows (visual/pipelines/)


Status

Visual artifacts are informative, optional, and non-normative.

Последнее обновление