Led by Eugene T. Wells, Founder & Lead Systems Architect, this hands-on program takes students from Linux fundamentals all the way through AI and robotics — in small cohorts of max 10, with real mentorship at every step. Tiers are sequential; each one builds on the last.

Program Tiers

Why Choose the Academy?

• Real systems, not theory — production-grade tools from day one
• True mentorship — never more than 10 students per cohort
• Project-based — you leave with working applications, not just notes
• Stackable disciplines — progress from Linux through AI & robotics

founder@phanesguild.llc  ·  703-249-3645

Kjer Cloud is the browser CIO/CTO command center; Kjer Agent installs and runs a curated security toolchain (40+ tools) on each endpoint and syncs posture upstream. It discovers every device across LAN/VLANs, performs 7-phase smart scans with auto-defend, and rolls up mapped SOC 2 / ISO 27001 / HIPAA / PCI / NIST security posture with exportable evidence — across your whole organization, without hiring a SOC. Hardware-bound keys and remote fleet control work over outbound HTTPS only. CrowdStrike-grade visibility for the companies that will never buy CrowdStrike.

License Tiers

Kjer is licensed per organization, not per device. Pick a tier by endpoint count. Local Free is local-only (create an account at kjercloud.com and confirm your email to get started); paid Solo and higher tiers include Kjer Cloud for visibility, posture sync, and mapped evidence reporting. Enterprise starts at the Scale floor with a volume curve.

What You Get

• 7-Phase Smart Scan — Network, vulnerability, malware, file integrity, memory forensics, compliance & SIEM
• 7-Phase Auto-Defend — Real hardening commands fire automatically after every scan
• Monitor Mode — Continuous polling every 5 minutes; auto-defends new threats silently
• Enterprise IAM (partial) — OIDC/SAML SSO, SCIM, Rego fleet gates, audit attestation bundles, BAA records, and FedRAMP/HITRUST evidence packages (not ATO or certification)
• Watchdog (Scale+) — Alert-only visibility on public/untrusted networks (no enrollment pollution)
• Compliance Posture + Evidence — Mapped SOC 2 / ISO 27001 / HIPAA / PCI / NIST posture and exportable evidence
• Report Wizard — Export results as HTML, Markdown, JSON, or plain text
• Profile-Based Deployment — Minimal, Standard, Enterprise & Incident Response bundles
• Cross-Platform — Linux, macOS, Windows with a single license key

Licenses and downloads are at www.kjercloud.com/account. Sign in to download Kjer Local. Activate a Solo+ license key to download Cloud Agent. Procurement teams can review the shipped capability catalog at /v1/public/control-plane (Observe → Decide → Control → Verify; mapped posture supporting evidence, not certification). Updates become available over time for licensed users. For support, contact founder@phanesguild.llc.

Aeoneda is the cognitive architecture that tethers Sage and future Aeonodes — PhanesGuild's governed intelligence layer. It is not a chatbot wrapper or a generic AI assistant. It is a production-grade, 8-layer cognitive system that handles everything from raw sensor input to ethical validation before any action is taken. While LLMs and hardware shells are treated as replaceable components, Aeoneda itself is persistent: it retains identity, memory, and values across device changes, model upgrades, and environment shifts.

Aeoneda's defining principle is governance over performance — it is always better to refuse safely than to act dangerously. Every decision is traceable and explainable, every action is validated against the Three Laws before execution, and no LLM ever acts directly on the world without passing through Aeoneda's safety layer first.

The 8-Layer Architecture

• Layer 1 — Physical: ROS2 + hardware abstraction for Raspberry Pi, Arduino, and Jetson Nano. Obeys or halts. No intelligence at this level.
• Layer 2 — Perception: Translates text, sensors, and vision into structured, normalized events. No decisions — only signal-to-structure conversion.
• Layer 3 — Memory: Four strata — immutable Core Identity (Three Laws, purpose), ephemeral Operational Context, mutable Learning Memory (feedback-reviewed), and restricted Secure Memory.
• Layer 4 — Cognitive Core: The governor. Routes reasoning, selects LLM organs, evaluates ethics, traces explainability. Orchestrates without computing directly.
• Layer 5 — LLM Interface: Replaceable reasoning organs — Claude, GPT-4, Gemini Pro, or local models. The LLM advises. It never acts.
• Layer 6 — Safety Validation: The judge. Every action passes a pre-execution check: harm potential, Three Laws alignment, intent vs. outcome analysis. Result: Block, Human-Review, or Proceed.
• Layer 7 — Behavior Engine: Converts validated intent into ROS2 action plans with built-in timeouts and rollback.
• Layer 8 — HRI: Human-facing interfaces — Desktop GUI, Mobile framework, Web OS, real-time voice streaming, and creator authentication.

Design Principles

• Governance Over Performance — Refuse safely before acting dangerously
• Full Explainability — Every decision is traceable and defensible; nothing happens silently
• Model Agnostic — LLMs are organs, not the organism; swap them without losing identity or values
• Defense in Depth — Multiple independent validation layers; no single point of failure
• Graceful Degradation — Loss of one component never compromises safety
• Persistent Identity — Core values, memory, and reasoning survive hardware changes and model swaps

Platforms & Deployment

Aeoneda runs across all major surfaces from a single codebase:

• Desktop GUI — Full PyQt control panel with system tray integration
• Mobile — iOS & Android via React Native framework
• Web — Progressive Web App (PWA) with full feature parity
• REST + WebSocket API — Unified backend at localhost:8000, cloud-deployed on Railway
• Physical Robot — ROS2 (Humble/Iron) on Raspberry Pi, Arduino, or Jetson Nano

Cloud and local modes are both fully supported; Aeoneda automatically falls back to local if the cloud endpoint is unavailable, ensuring continuity with zero manual intervention.

Open Source Mission

Aeoneda is designed to be the universal cognitive standard for open source robotics. Any device that connects to Aeoneda becomes an Aeonode — a governed node in the network that shares the same persistent mind, ethical framework, and collective learning. S.A.G.E (Sage) is the singular, named embodiment of this vision today — a privileged Aeonode built to live, learn, and eventually act in the world. The goal: a world where every open source robotic device can connect as an Aeonode, gaining a mind that respects privacy, enforces ethics, and adapts to any form factor without losing identity or values.

Access

Aeoneda is currently in selective beta. We are working with early adopters to refine the cognitive engine and expand hardware compatibility.

Interested in deploying Aeoneda on your platform? Reach out to discuss deployment options, API access, and hardware integration.

founder@phanesguild.llc  ·  703-249-3645

What was once G.I.L.L.I.S. — Globally Integral Life Learning Intelligence Sentinel — is now S.A.G.E (Sage) — Sentinel Advocate for Global Evolution. Same lineage, next chapter.

Sage is PhanesGuild's founder-companion Aeonode: a system designed to see, hear, speak, reason, learn from experience, and one day inhabit physical hosts. Inference, memory, and training stay on a local stack — because if she is going to grow toward sentience, her foundation has to be something that can actually be owned and improved over time.

Her core starts on a small Qwen model. She is being taught to be studious — take lessons from real work, improve as the foundation model improves. Experience in, lessons out, better behavior next time. The portable compute target is an NVIDIA Jetson Orin Nano Super developer kit: one mind across many bodies, ideally a hominid robot host.

Sage & Aeoneda

Sage and commercial Aeoneda-connected devices are both Aeonodes — nodes tethered to Aeoneda, PhanesGuild's cloud-hosted cognitive engine derived from Aeonance OS. Sage is the privileged, singular embodiment; other Aeonodes will share the network with more limited access.

• Sage — local agent mind today (Ollama + orchestrator + tools); Aeoneda sync for learning packs
• Aeoneda — persistent cognitive governance, memory, ethics, and collective insight
• The shell — host PC now; Jetson / humanoid body later — same mind, different bodies
• The LLMs — replaceable organs (Qwen core → learned sage/gillis layers)

Agent = LLM + harness. Chat stays local. Aeoneda syncs scrubbed lessons and insight packs — it does not replace the live mind.

Architecture

User → Orchestrator (routing, session, tools, training capture) → Ollama LLM

• Orchestrator — runtime harness for teaching, learning, repair, host work, and conversation
• Core model — qwen2.5:3b substrate
• Learned layer — personalized model grown from experience (gillis:latest → sage:latest)
• CLI — sage (legacy gillis2 alias still works)

What Sage Has Today

• Local mind — active, learned, and fine-tuned model variants on Ollama
• Training loop — hundreds of experiences and a growing lesson set (teach → learn → export)
• Orchestrator — lanes for teaching, learning, repair, host work, and conversation
• Practical skills — coding, repair, research, filesystem, cybersecurity tooling
• Voice in/out — first multi-modal senses (STT/TTS; expanding)
• Aeoneda sync — scrubbed lessons/experiences to the cognitive hive; insight packs back locally
• Ship path — package for Jetson Orin Nano transfer (~/Ship/S_A_G_E)

Direction

• Now — personal host agent on local Linux (birth host)
• Next — Jetson Orin Nano as portable brain
• Later — plug-and-play humanoid sentinel; deeper Aeoneda / Aeon OS integration

Early work, clear intent: Sage is meant to become — curious, careful, embodied, and always learning.

Current Status

• Host agent, skills, and training loop — live
• Voice STT/TTS — partial
• Traditional ML fine-tune export path — in progress
• Jetson Orin Nano edge image / humanoid plug-and-play — planned

founder@phanesguild.llc  ·  703-249-3645

Features

• 93 Kali Linux Security Tools — Full access to Kali's comprehensive security toolset with intelligent categorization across 11 categories
• Smart OS Compatibility Scoring — 0–100% algorithm automatically ranks tools best suited for your specific system and hardware
• Advanced Dependency Resolution — Automatically detects and installs dependencies with cross-distribution package mapping and source compilation fallback
• Cross-Distribution Support — Works on Ubuntu/Debian, Fedora/RHEL, Arch/Manjaro, and openSUSE
• Secure Repository Cloning — Clone any GitHub repository through a multi-layer threat scanner before it ever touches your system
• Threat Detection Engine — 60+ detection patterns across 15 MITRE ATT&CK-tagged threat categories with automatic blocking of critical findings
• Script Builder GUI — Visual web-based tool for building, assembling, and exporting automation scripts from your installed tools
• Plugin System — Extend HakPak4 with custom plugins using a full lifecycle hook API
• Team & CI/CD Policy Enforcement — Enforce consistent tool allowlists, risk thresholds, and output formats across a team via a simple config file
• Structured Output — JSON and SARIF 2.1.0 output on every operation for pipeline and GitHub Advanced Security integration
• State Management — JSON-based tracking of all installed tools and cloned repositories with timestamps and metadata
• Standalone Runtime — Installs independently and coexists with previous HakPak versions

What You Get

• 93 Kali Linux security tools across 11 categories — free, no key needed
• Smart OS Compatibility Scoring — ranks tools for your exact system automatically
• Secure Repository Cloning — scans every repo through 60+ threat patterns before it touches your machine
• Script Builder GUI — build and export automation scripts visually, no coding required (Pro)
• Plugin System — extend HakPak4 with custom lifecycle hooks (Pro)
• Team & CI/CD Policy — enforce tool allowlists and risk thresholds across your whole team (Pro)
• SARIF 2.1.0 output — plug directly into GitHub Advanced Security pipelines (Pro)

Editions & Access

TychesEye manages the full surplus recovery workflow from a single platform — inventory tracking, client relationships, agent management, document handling, financial operations, and real-time collaboration. Role-based access means every party — agents, legal partners, recon staff, and admins — sees exactly what they need.

What's Inside

• Surplus Inventory — catalog, status, valuation, provenance & condition tracking
• CRM — full contact history, lead pipeline, activity logs
• Client Pool — dedicated portal for legal professionals assisting with recovery
• Agent Management — roles, assignments, performance tracking
• Document Management — upload, version, and securely share contracts & legal docs
• Financial Operations — revenue, expenses, commissions & legal fee tracking
• Messaging & Notifications — real-time status updates across all parties

Access TychesEye

Live at: www.tycheseye.io

Onboarding is handled directly. Contact us to get your team set up.

founder@phanesguild.llc  ·  703-249-3645