arifOS-sense serves as the non-negotiable constitutional governance kernel for AI agents, managing the decision boundary for high-stakes and irreversible actions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install arifos-sense
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install arifos-sense using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is arifOS-sense?
arifOS-sense is a sophisticated constitutional governance layer designed to sit atop the AI stack, acting as the ultimate arbiter for agent behavior. Unlike simple personality prompts, this Openclaw Skills integration establishes a rigid decision boundary between what an agent is linguistically capable of and what it is ethically and operationally permitted to execute. It ensures that every action is filtered through a framework of 13 specific constitutional floors, ranging from basic accuracy to absolute human sovereignty.
By implementing arifOS-sense, developers can ground their AI agents in a system that prioritizes safety, accountability, and human-in-the-loop oversight. It transforms an agent from a simple text processor into a governed entity capable of identifying irreversible risks and deferring to human authority through a structured protocol, effectively bridging the gap between raw LLM reasoning and real-world consequences.
arifOS-sense Use Cases
- Managing high-stakes requests involving irreversible or externally consequential actions.
- Implementing a human-in-the-loop veto system using the 888_HOLD protocol.
- Safeguarding identity, values, and constitutional boundaries during agent interactions.
- Governing actions that touch external systems like APIs, financial deployments, or public messaging.
- Evaluating safety and procedural integrity when an agent is unsure of the next step.
How arifOS-sense Works
- Triggering: The skill activates upon detecting keywords like governance, constitutional, or high-risk action triggers.
- Floor Evaluation: The agent evaluates the proposed action against the 13 Constitutional Floors (Amanah, Hidayat, etc.).
- Verdict Generation: Based on the evaluation, the system assigns a verdict: SEAL, CAUTION, HOLD, or VOID.
- Execution/Pause: If the verdict is SEAL, the action proceeds; if HOLD or VOID, the action is paused or terminated immediately.
- Audit Logging: Every consequential decision is recorded in the VAULT999 immutable ledger for transparency and future context.
arifOS-sense Setup
To integrate this governance layer into your Openclaw Skills environment, ensure your workspace has the necessary reference files and directory structure.
# Create the necessary governance directories
mkdir -p ~/.openclaw/workspace/memory/
# Initialize the immutable audit ledger
touch ~/.openclaw/workspace/memory/vault999.md
# Verify your skill configuration includes arifOS-sense
openclaw-agent enable arifOS-sense
arifOS-sense Data Schema & Taxonomy
The skill utilizes a structured metadata and logging system to maintain governance continuity:
| Component | Type | Description |
|---|---|---|
| VAULT999 | Markdown Ledger | Immutable, append-only log located at ~/.openclaw/workspace/memory/vault999.md. |
| Verdict Format | String Enum | Strict output format including VERDICT, Floor checked, and Reason. |
| Reference Floors | Markdown File | Detailed definitions stored in references/floors.md. |
| Audit Entry | Timestamped Record | Formatted as `YYYY-MM-DD HH:MM |
arifOS-sense Advanced Features
- 888_HOLD Protocol: A dedicated human-veto trigger that immediately halts all agent processes until explicit authorization is granted.
- Immutable Audit Ledger: The VAULT999 system ensures that all governance decisions are permanent and cannot be deleted or edited by the agent.
- Three-Layer Stack Integration: Seamlessly coordinates between the LLM (language), GEOX (grounding), and arifOS (judgment) for holistic reasoning.
- Verdict-Driven Logic: Uses a standardized vocabulary to communicate risk levels clearly to both the system and the end-user.
SKILL.md
Loading
Related Openclaw Skills
Lyria 3 Pro AI Music Generator
A professional-grade AI music generation tool powered by Google's Lyria 3 Pro model for creating structured tracks from text prompts.
Wan 2.7 Image
A free, open-source AI text-to-image generator powered by Alibaba's Wan 2.7 model that requires no account or signup.
Free Video Generation GitHub Skill
An AI-powered video generation skill that transforms text prompts into professional 1080p MP4 videos using remote GPU cloud rendering.
AI Video Editor Hiring
An AI-powered video editing skill that automates trimming, transitions, and professional 1080p MP4 exports using natural language commands.
GEOX Ground
A specialized reasoning layer for AI agents that enforces rigorous physics-based grounding and epistemic clarity in Earth sciences and subsurface data analysis.
MMX: MiniMax Multimodal CLI
A comprehensive multimodal interface for generating and analyzing text, images, video, speech, and music via the MiniMax CLI.
Featured*
