Skip to content

Architecture & Flows

High-level design of the AI Hub: how components connect, how data flows between them, and how the repository maintains itself over time. Updated for early January 2027 SOTA standards (incorporating FastMCP 3.1, Claude 5.1, GPT-5.5/5.6, Gemini 4.0 Pro, and multi-agent KnowledgeOps).

Contents

Document What it covers
Component Map Full inventory of services and their relationships — the definitive "what talks to what" map
Automation Flows Detailed sequence and data-flow diagrams for key automation workflows
Infrastructure Hardware topology, network layout, k3s clusters, and resource allocation decisions
SSH Execution Patterns Secure orchestration of remote commands across TrueNAS, Pi, and MacBook
MCP Patterns Architecture for tool-calling via Model Context Protocol (FastMCP 3.1)
Automated Contributions How Google Jules, digest workflows, and quality gates keep the repo self-improving
Multi-Agent KnowledgeOps Governance contract, role model, and CI gates for scalable multi-agent documentation growth
Prompt Catalogue Reference library of production prompts used across automation workflows
Text-to-SQL Architecture Layered multi-agent pipeline for natural language data querying

System at a Glance

flowchart TD
    subgraph Ingest
        A[Daily Digest / External Intake] --> B[Intake Bridge]
        B --> C[docs/new-sources/]
    end
    subgraph Ralph-loop
        C --> D[Jules Issue]
        D --> E[Jules Execution\n(a) Work (b) Link (c) Decompose]
        E --> F[Jules PR / Commit]
    end
    subgraph Quality Gates
        F --> G[Audit Docs Quality]
        G --> H[Check Docs Contract]
        H --> I[Check Catalog Consistency]
        I --> J[Main Branch]
    end
    subgraph Deployment
        J --> K[MkDocs Build]
        K --> L[GitHub Pages / Cloudflare Pages]
    end

Sources / References

Contribution Metadata

  • Last reviewed: 2027-01-07
  • Confidence: high