moses-modes

Apply mode-driven governance constraints to agent prompts from a state file.

1|Updated Mar 12, 2026

Search Tags:#agents #governance #constraints #prompts #openclaw #modes

Ready to run

npx skills add https://github.com/SunrisesIllNeverSee/moses-claw-gov --skill moses-modes

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: moses-modes
Source: https://github.com/SunrisesIllNeverSee/moses-claw-gov/tree/main/skills/moses-modes
Command: npx skills add https://github.com/SunrisesIllNeverSee/moses-claw-gov --skill moses-modes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

MO§ES™ Governance Modes injects behavioral constraints from the active mode into all agent prompts, reducing the risk of unsafe or non-compliant actions.

Core Features & Use Cases

Active-mode enforcement: automatically apply constraints (high-security, high-integrity, creative, etc.) before every action.
Context-aware: reads state from ~/.openclaw/governance/state.json before executing prompts.
Use Case: ensure safe, auditable prompts in financial operations, risk management, and regulated environments.

Quick Start

Set the active governance mode via the operator command (e.g., /govern high-security).

Frequently Asked Questions about moses-modes

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I enforce AI governance constraints across all agent prompts?▼

AI governance modes inject behavioral constraints from an active mode into all agent prompts, reducing unsafe or non-compliant actions. The system reads constraints from a state file and applies them before execution.

How do I apply high-security constraints to financial prompts?▼

Apply high-security constraints to financial prompts by setting the active governance mode to high-security. This mode automatically gates actions, logs decisions, and provides audit-ready outputs for regulated environments.

What governance modes are available for AI agents?▼

Available governance modes include high-security, high-integrity, creative, research, self growth, problem solving, I Don't Know What To Do, and None. Each mode injects specific behavioral constraints into agent prompts.

Does the governance mode system require a specific state file format?▼

Yes, the system requires reading state from ~/.openclaw/governance/state.json before executing prompts. This file stores the active mode that determines which behavioral constraints are applied to your agents.

Can I use mode-driven constraints for analytical and learning tasks?▼

Yes, you can use mode-driven constraints for analytical and learning tasks. The system supports applying different modes like research or self growth to gate actions and ensure compliant behavior across these domains.

What is the best way to switch between integrity and creative modes for agents?▼

Switch between integrity and creative modes by using the operator command to set the active governance mode. This updates the state file, automatically applying the new constraints to all subsequent agent prompts.