GitHub – RunTimeAdmin/ai-agent-killswitch

AI Agent Safety Control – Emergency kill switch and guardrails for AI agents

$KILLSWITCH is a comprehensive safety ecosystem for AI agents, powered by Runtime Fence technology. Think of it as a kill switch for your AI – instantly stop any agent, block dangerous actions, and monitor everything in real-time.

🌐 Live Demo: killswitch.protocol14019.com

• 🔴 Kill Switch – Instantly stop any AI agent with one click
• 🚫 Action Blocking – Define what actions agents cannot take
• 🛡️ Target Protection – Block access to sensitive files, APIs, or systems
• 💰 Spending Limits – Control how much an agent can spend
• 📊 Risk Scoring – Automatic risk assessment of every action
• 📝 Audit Logging – Complete trail of all agent activity
• 📧 Email/SMS Alerts – Get notified of suspicious behavior
• 🖥️ Cross-Platform – Windows, macOS, and Linux support

• 🔒 Runtime Fence – Real-time action validation and monitoring
• ☠️ Kill Switch – SIGTERM → SIGKILL emergency termination
• 📝 Audit Logging – Complete cryptographic audit trail
• ⚡ Sub-Second Response – Kill signals under 100ms
• 🛡️ SPIFFE Identity – Cryptographic workload identity
• 🗳️ Token Governance – Decentralized oversight with $KILLSWITCH voting

pip install killswitch-agent

Or clone and install:

git clone https://github.com/RunTimeAdmin/ai-agent-killswitch.git
cd ai-agent-killswitch/packages/python
pip install -e .

from runtime_fence import RuntimeFence, FenceConfig

# Create a fence
fence = RuntimeFence(FenceConfig(
    agent_id="my-agent",
    blocked_actions=["delete", "exec", "sudo"],
    blocked_targets=[".env", "production", "wallet"],
    spending_limit=100.0
))

# Validate an action
result = fence.validate("read", "document.txt")
if result.allowed:
    # Proceed with action
    pass
else:
    print(f"Blocked: {result.reasons}")

# Kill switch
fence.kill("Emergency stop")

@fence.wrap_function("api_call", "external_service")
def call_external_api(data):
    return requests.post("https://api.example.com", json=data)

# Now the function goes through the fence automatically
call_external_api({"key": "value"})

chmod +x install_fence.sh
./install_fence.sh

Look for the shield icon in your system tray. Right-click for options.

fence version --check    # Check for updates
fence update             # Upgrade to latest version
fence status             # Show fence status
fence scan               # Detect AI agents on your system
fence test               # Run quick validation test
fence start              # Launch tray app

$KILLSWITCH includes presets for common AI agents:

┌─────────────────────────────────────────────────────────┐
│                     Your AI Agent                        │
└─────────────────────────┬───────────────────────────────┘
                          │
                          ▼
┌─────────────────────────────────────────────────────────┐
│                   $KILLSWITCH                            │
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐     │
│  │ Validator   │  │ Risk Scorer │  │ Kill Switch │     │
│  └─────────────┘  └─────────────┘  └─────────────┘     │
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐     │
│  │ Audit Log   │  │ Alerts      │  │ Settings    │     │
│  └─────────────┘  └─────────────┘  └─────────────┘     │
└─────────────────────────┬───────────────────────────────┘
                          │
                          ▼ (if allowed)
┌─────────────────────────────────────────────────────────┐
│                   External World                         │
│         (APIs, Files, Databases, Network)               │
└─────────────────────────────────────────────────────────┘

# Using JWT token
curl -H "Authorization: Bearer " https://api.runtimefence.com/api/runtime/status

# Using API key
curl -H "X-API-Key: ks_xxxxx" https://api.runtimefence.com/api/runtime/status

# API
KILLSWITCH_API_URL=http://localhost:3001
KILLSWITCH_API_KEY=your_api_key

# Alerts
SMTP_HOST=smtp.gmail.com
SMTP_PORT=587
SMTP_USER=your@email.com
SMTP_PASSWORD=your_password
ALERT_TO_EMAILS=admin@company.com

# SMS (Twilio)
TWILIO_SID=your_sid
TWILIO_TOKEN=your_token
TWILIO_FROM=+1234567890
ALERT_SMS_NUMBERS=+1234567890

Access the web dashboard at http://localhost:3000/settings to configure:

• Agent presets (Coding, Email, Data, Web, Autonomous)
• Blocked actions and targets
• Spending limits
• Risk thresholds
• Auto-kill settings
• Email/SMS alerts

# Clone repo
git clone https://github.com/RunTimeAdmin/ai-agent-killswitch.git
cd ai-agent-killswitch

# Install dependencies
npm install

# Start development servers
npm run dev

# Run tests
npm test

ai-agent-killswitch/
├── apps/
│   └── web/              # Next.js dashboard
├── packages/
│   ├── core/             # Kill switch engine
│   ├── sdk/              # TypeScript SDK
│   ├── cli/              # Command-line tools
│   └── python/           # Python fence wrapper
├── services/
│   └── api/              # Express REST API
└── docs/                 # Documentation

Built by RunTimeAdmin | David Cooper | CCIE #14019

Related Projects:

Why This Matters:

“When AI agents go rogue in Kubernetes, you need a kill switch. Not a button. A network-level containment system.”

As Fortune, Palo Alto Networks, and Andrej Karpathy warn of the emerging AI security crisis, we’re building real solutions—not just launching tokens.

$KILLSWITCH provides the guardrails that prevent AI agents from causing catastrophic damage, while the token enables community governance and sustainable development of the ecosystem.

MIT License – see LICENSE for details.

# Stop an agent from deleting files
from runtime_fence import RuntimeFence

fence = RuntimeFence()
result = fence.validate("delete", "production_database.sql")
# Returns: {"allowed": False, "reasons": ["Action 'delete' is blocked"]}

# Get risk score for a transaction
result = fence.validate("transfer", "0x...", amount=1000000)
# Returns: {"allowed": False, "risk_score": 95, "reasons": ["High-value transfer blocked"]}

# Emergency stop
fence.kill("Suspicious activity detected")
# All agent operations halted immediately

🛡️ Protect your AI. Before it protects itself.

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

Contract: 56o8um92XU8QMr1FsSj4nkExEkgKe56PBTAMqCAzmoon

Buy on Jupiter

• Runtime Fence engine
• Python and TypeScript SDKs
• REST API
• Web dashboard

• USD subscriptions (Stripe)
• Crypto payments (SOL/USDC)
• $KILLSWITCH token utility
• Token-weighted governance
• Usage-based tier limits

• Mobile app (iOS/Android)
• Third-party integrations
• Plugin marketplace
• Enterprise features
• Multi-agent orchestration

$KILLSWITCH – Because every AI needs an off switch.