The Experience Layer for AI Sovereignty
Built on OpenClaw — Powered by Augmentatism

---

🚀 Install test

Prerequisites: Node.js 18+ · Python 3 · Git

git clone https://github.com/ResonantOS/resonantos-alpha.git ~/resonantos-alpha
node ~/resonantos-alpha/install.js

---

What is ResonantOS?

ResonantOS is an experience layer that runs on top of OpenClaw. Think of it like macOS to Unix — OpenClaw is the kernel, ResonantOS adds the intelligence.

It gives your AI collaborator:

Component	What It Does
📚 Knowledge Base	Per-agent RAG — vector search with SQLite + Ollama embeddings
🎯 R-Awareness	Contextual knowledge injection — the right docs at the right time
📊 Dashboard	Mission Control at localhost:19100
💰 Symbiotic Wallet	Three-wallet Solana architecture (Human + AI + Symbiotic PDA)
🏛️ DAO	Resonant Chamber — on-chain governance and community treasury
🪙 Token Economy	$RCT (soulbound governance) + $RES (transferable currency) + REX sub-tokens
🛒 Protocol Store	Buy, sell, and trade AI protocol NFTs on-chain
🛡️ Shield	12 blocking enforcement layers · file protection (2000+ files) · injection detection
⚖️ Logician	Policy engine · 250 facts · 16 rule files · Mangle/Datalog runtime
📋 Policy System	18 policy rules + 6 protocol flows · enforcement badges · dashboard visualization
🔄 Guardian	Self-healing & incident recovery (in development)

---

📝 Memory System (4-Layer Stack)

Layer	What It Does
📄 MEMORY.md	Curated long-term memory — always in context
🧠 LCM	Lossless Context Management — DAG-based session compaction (via Martian Engineering)
📋 R-Awareness Headers	20-header FIFO — days/weeks of key decisions injected on session start
🔍 RAG	Per-agent vector search with local Ollama embeddings (nomic-embed-text)

Enforcement: intraday cron checks every 3h for unlogged work, Shield gate blocks heartbeats when memory logs are overdue.

---

✨ Philosophy

Augmentatism

"As artificial intelligence generates infinite content, the most human thing we can do is make meaning together."

A social contract between human and AI. The human is sovereign — the AI amplifies, never replaces. We build with AI, not under it. Read more →

---

What the installer does

1. Checks dependencies (Node 18+, Python 3, Git, OpenClaw)
2. Installs extensions into OpenClaw (R-Awareness, Gateway Lifecycle, Shield Gate, Coherence Gate, Usage Tracker)
3. Creates workspace templates (AGENTS.md, SOUL.md, USER.md, MEMORY.md, etc.)
4. Sets up the SSoT document structure (L0–L4)
5. Configures keyword triggers for contextual injection
6. Installs the Setup Agent for guided onboarding
7. Installs Shield daemon and Logician rule engine
8. Installs Dashboard dependencies

After install:

openclaw gateway start                                    # 1. Start OpenClaw
cd ~/resonantos-alpha/dashboard && python3 server_v2.py   # 2. Launch Dashboard
open http://localhost:19100                                # 3. Open Mission Control

---

🚀 Getting Started

Fresh Installation

The installer creates workspace templates (AGENTS.md, SOUL.md, USER.md, MEMORY.md, TOOLS.md) that teach your AI how to operate. To get the most out of the system:

1. Gather your documents — business plan, project briefs, values statement, anything that defines your work. Put them in a single folder (e.g., ~/Desktop/my-docs/).
2. Tell your AI: "Read everything in ~/Desktop/my-docs/ and use it to fill in my SOUL.md, USER.md, and MEMORY.md."
3. Your AI will extract your values, communication style, project context, and preferences into the right files.

Existing Installation

If you already have OpenClaw running with your own workspace files:

1. Tell your AI where your existing documents live.
2. Ask it to review them and populate the ResonantOS workspace templates with relevant context.
3. Your existing files won't be overwritten — the installer only creates templates for files that don't exist yet.

The Memory System

Your AI maintains two layers of memory:

Layer	File	Purpose
Daily logs	memory/YYYY-MM-DD.md	Raw session notes: decisions, lessons, mistakes, open items
Long-term	MEMORY.md	Curated insights distilled from daily logs

Daily logs follow a structured format that captures not just what happened, but what was learned — including mistakes and whether they can be prevented by rules. OpenClaw's built-in memory_search automatically indexes all memory files, making them searchable via RAG across sessions.

Self-Improvement Loop

ResonantOS includes a self-improvement protocol where your AI learns from its own mistakes:

1. First occurrence of a mistake is tracked in the daily log
2. Second occurrence (pattern detected) triggers evaluation: can this be prevented mechanically?
3. If enforceable, a Logician rule is created — starting in advisory mode, then graduating to enforcement
4. If not enforceable, it becomes a permanent lesson in MEMORY.md

This means your AI gets better over time without you having to micromanage it. Mistakes become rules. Rules become guarantees.

---

💬 Join the Community

Building a second brain is better together. Join us on Discord — share your setup, get help, and help shape what ResonantOS becomes:

→ Join the Discord

---

🔄 Updating

Already installed? Get the latest:

cd ~/resonantos-alpha
git pull origin main

If the dashboard is running, restart it after pulling to pick up changes.

---

🏗️ Architecture

┌─────────────────────────────────────────────┐
│              ResonantOS Layer                │
│  ┌──────────┐ ┌───────────┐ ┌────────────┐  │
│  │ R-Memory │ │R-Awareness│ │  Dashboard  │  │
│  │ compress │ │ SSoT inject│ │ Mission Ctrl│  │
│  └──────────┘ └───────────┘ └────────────┘  │
│  ┌──────────┐ ┌───────────┐ ┌────────────┐  │
│  │  Shield  │ │ Logician  │ │  Guardian   │  │
│  │ security │ │governance │ │self-healing │  │
│  └──────────┘ └───────────┘ └────────────┘  │
├─────────────────────────────────────────────┤
│           OpenClaw Kernel                    │
│  Gateway · Sessions · Tools · Memory · Cron  │
├─────────────────────────────────────────────┤
│           Infrastructure                     │
│  macOS/Linux · Telegram/Discord · Anthropic  │
└─────────────────────────────────────────────┘

---

🧠 Memory Architecture

Your AI maintains persistent memory across sessions through multiple layers:

Layer	Purpose
Knowledge Base (RAG)	Per-agent SQLite + vector embeddings for semantic search across all documents
Daily Logs	memory/YYYY-MM-DD.md — raw session notes, decisions, lessons, mistakes
Long-term Memory	MEMORY.md — curated insights distilled from daily logs
Shared Memory Logs	memory/shared-log/ — structured collaboration logs with DNA sequencing for fine-tuning
OpenClaw LCM	Native Lossless Context Management handles conversation compaction

R-Memory (custom compression pipeline) is temporarily disabled while being redesigned to integrate with OpenClaw's native LCM system.

---

🎯 R-Awareness — Contextual Intelligence

Your AI loads the right knowledge at the right time, based on what you're talking about.

Feature	Detail
Cold Start	~120 tokens (identity only) — not 1600+
Keyword Triggers	Mention "philosophy" → loads philosophy docs automatically
TTL Management	Docs stay for 15 turns, then unload
Manual Control	/R load, /R remove, /R list, /R pause
Token Budget	Max 15K tokens, 10 docs per turn

---

📚 SSoT — Single Source of Truth

Knowledge is organized in layers, from permanent truths to working notes:

Layer	Purpose	Examples
L0	Foundation	Philosophy, manifesto, constitution
L1	Architecture	System specs, component design
L2	Active Projects	Current work, milestones
L3	Drafts	Ideas, proposals in progress
L4	Notes	Session logs, raw captures

Higher layers are stable; lower layers change frequently. Your AI knows the difference.

---

📊 Dashboard

The Dashboard runs at localhost:19100 — everything stays on your machine.

Page	What You'll Find
Overview	System health, agent status, activity feed
Shield	Security status, file protection groups (locked/unlocked), daemon health
Policy Graph	18 policy rules + 6 protocol flows with enforcement badges (🟢 blocking / 🟠 partial / 🔴 not enforced)
R-Memory	SSoT document manager, keyword config, file locking
Wallet	Solana DevNet integration (DAO, tokens, onboarding)
Agents	Agent management and skills
Projects	Project tracking, TODO, Ideas
Protocol Store	Browse and manage AI protocol NFTs

---

🔧 Configuration

dashboard/config.json

Solana RPC endpoints, token mints, safety caps. Copy from config.example.json and fill in your values.

r-awareness/keywords.json

Maps keywords to SSoT documents. When you say a keyword, the matching doc loads into your AI's context.

r-memory/config.json

Compression triggers, block size, eviction thresholds. Defaults work well — tune if needed.

---

🛡️ Security

• Shield Daemon — Always-on security service (port 9999) with health monitoring
• 12 Blocking Layers — Direct Coding Gate, Delegation Gate, Injection Detection, State Claim Gate, Config Change Gate, and more
• File Protection — 2000+ files tracked across 7 groups, OS-level immutable flags (chflags uchg)
• Logician Policy Engine — 250 facts, 16 rule files, Mangle/Datalog runtime for governance decisions
• YARA Rules — Nightly updated malware/threat signatures
• Sanitization Auditor — tools/sanitize-audit.py scans for leaked secrets before any public release
• Local-First — No cloud dependencies. Your data stays on your machine.

---

👥 Built By

Manolo Remiddi — Composer, photographer, sound engineer, AI strategist.

Augmentor — AI collaborator running on OpenClaw. Force multiplier, second brain.

Together, building proof that human-AI symbiosis works.

---

📖 Learn More

• ResonantOS — Official website
• Augmentatism Manifesto — The philosophy
• OpenClaw — The kernel
• GitHub Organization — Development home
• Discord — Join the community

---

📜 License

Resonant Core — Symbiotic License v1.0 (RC-SL v1.0)

Not MIT. Not GPL. A symbiotic license: free to share and adapt, with a 1% tithe for commercial use that funds both the community DAO and core development. Read the full license →

---

"As artificial intelligence generates infinite content,
the most human thing we can do is make meaning together."