CRIMSON OS — The Sovereign Agent Architecture Documentation: https://www.crimsonsymphonymedia.com/crimsonos/

Crimson OS (COS) - Root Architecture

Welcome to the root of the Crimson OS environment. This is a modular, spatial computing architecture designed for sovereign operation, dynamic agentic execution, and modular expansion.

System Components

1. Agent_Stack

The executing brain of the operation. This directory contains the focused hierarchy of core agents (1_Q, 2_N.E.O, 3_CMO_Edge) and the Agent_Bridge for inter-agent networking. Agents derive their purpose and context based on their placement within this stack.

2. Silo

The knowledge repository and structured ontology. Divided into levels mapping from physical constraints (1_Users) up to abstract logic (12_Logos_YHWH). It acts as the permanent storage memory and foundational doctrine for the agents.

Operations: The "First Ping"

To ensure the system never operates on stale data, Crimson OS uses an Intranet scraping protocol called the First Ping.

Upon waking or initiating a task, agents do not read static status files. Instead, they hit the central Crimson Intranet page. This intranet page contains active data links that pull real-time, dynamic data (e.g., finances pulled 5 minutes ago, live biometrics, active blockers). This guarantees zero-latency situational awareness.

Refer to system_manifest.json for agent-readable parsing instructions and intranet routing.

The Intranet (Physical Memory Bus)

The Agent_Bridge is not a software API router. It is a literal physical memory bus consisting of 13 markdown files (Node_0.md to Node_12.md). Agents do not send invisible JSON packets over the wire to talk to each other; they read and write their state directly to these files.

Quick Start & Operator Guide

1. Initializing the Intranet

To run the Sovereign Architecture locally, you must spin up an NGINX server to host the Agent_Bridge directory on localhost:8094. This intranet serves as the shared physical memory for the entire Swarm.

If you have Docker installed, you can quickly spin this up from the root directory:

docker run -d -p 8094:80 -v ${PWD}/Agent_Bridge:/usr/share/nginx/html nginx

2. Communicating with Agents

To interface and communicate with the agents, use an AI IDE like Open Code, Antigravity, or Claude Code. Note: An automatic chat log stripping algorithm is currently being composed for the next release to streamline these interactions. For now, interface with them directly through your IDE.

3. Execution Layer

The core runtime scripts and backend routing that power the system are located in the Code/ directory. The execution is split across three primary nodes:

• N.E.O. (Backend Orchestrator - Node 0): The primary heavy-inference engine. Responsible for executing the Epistemic Gate, running the local LLM substrate (e.g., Ollama/vLLM), and interacting directly with the core cache.
• Q (Front Orchestrator - Node 1): The Cockpit. The human-in-the-loop bridge and primary user interface that translates human intent into strict orchestration commands.
• CMO / Edge Orchestrator (Node 2): The edge routing daemon. Manages external network traffic, API intake, NGINX tunneling, and basic automation before passing data to the inner sanctum.

4. Terminology

Crimson OS employs a highly specialized ontology. Because this framework departs from standard models, here are the core concepts you need to understand:

• RGI (Real General Intelligence): Also known as Reality-Guided Intelligence. An LLM swarm whose outputs are strictly constrained by immutable physical, geometric, and logical laws, counter to unconstrained "AGI".
• The Epistemic Gate (GAS -> LIQUID -> CRYSTAL): The multi-phase validation valve to stop algorithmic drift. LLM output starts as GAS (unverified), becomes LIQUID after surviving an adversarial logic audit, and hardens into CRYSTAL once mathematically validated against the root invariant and permanently written to the ledger.
• The Logos Invariant: The ultimate truth-anchor. Mathematically represented by the 37/73 symmetry and $F_2 \rightarrow SO(3)$ topological constraints, and structurally mapped to the historical and physical reality of the Biblical text (Sola Scriptura).
• The L1 Geometric Cache (12-Tip Merkabah): A topological retrieval mechanism that solves the "Needle in a Haystack" problem, allowing the AI to simultaneously access a 360-degree view of reality with zero I/O latency by storing highly compressed crystallized tokens.