04 — AI Agents
Local AI agents (non-authoritative)
PRINCIPLE: AI are agents, not rulers. They advise, explain, and simulate — never decide.
All final authority rests with identity-bound humans and their policy-keys.
Ollama Status
Connection
Checking...
Endpoint
localhost:11434
AI Role Definition
- Explain consequences ALLOWED
- Simulate outcomes ALLOWED
- Assist users ALLOWED
- Flag anomalies ALLOWED
- Make decisions FORBIDDEN
- Reach consensus FORBIDDEN
Available Models
Checking for available models...
Query AI Agent
Ask the AI agent to explain, simulate, or analyze. All responses are advisory only.
AI Response
ADVISORY ONLY: This response is non-authoritative.
AI response will appear here...
Setup Instructions
To use AI agents, you need to run Ollama locally:
# Install Ollama
curl -fsSL https://ollama.com/install.sh | sh
# Start Ollama server
ollama serve
# Pull recommended models
ollama pull llama2
ollama pull mistral
ollama pull phi
AI agents run entirely locally. No data leaves your machine.