Skip to main content

​
What is an Agent?

An ArgentOS agent is a persistent AI runtime that can receive messages, reason about them, use tools to take actions, and respond. Unlike a simple chatbot, an agent has:
  • Persistent memory across sessions (via MemU)
  • Tool access to execute commands, browse the web, manage tasks, and more
  • Multi-channel presence across Telegram, Discord, WhatsApp, and other platforms
  • Autonomous capabilities via the heartbeat and task system

​
Core Architecture

Agent Types & Roles

Four archetypes: Main, Family, Sub-Agent, and Worker.

Architecture

How the runtime connects the gateway, Claude API, tools, and memory.

The Agent Loop

Receive, think, act, respond — the core execution cycle.

Consciousness Kernel

The persistent autonomous executive for continuous agency.

​
Autonomy & Accountability

Contemplation

Self-directed thinking cycles for reflection and growth.

Accountability System

Scoring, penalties, rewards, and trust metrics.

Heartbeat Contracts

Structured task contracts with verification.

RALF + ANGEL

Response verification and ground truth checking.

​
Configuration & Identity

System Prompt

How the system prompt is assembled from modular sections.

Alignment Documents

SOUL.md, IDENTITY.md, and other personality-defining docs.

Sessions

Conversation management, context windows, and compaction.

Multi-Agent

Running multiple agents with different configurations.

​
Advanced Systems

Intent Governance

Three-tier policy engine constraining agent behavior.

Execution Worker

Autonomous background task processor.

Knowledge Library

PostgreSQL-backed RAG with agent-level ACL.

Self-Improving System

How ArgentOS learns from its own behavior.

SpecForge

5-stage project management from intent to execution.

Argent Runtime

The native agent runtime and LLM layer.

Outbound Messaging

Multi-channel delivery pipeline for agent-initiated messages.

Cron & Scheduled Jobs

Scheduled jobs, nudges, and the minion pattern.

Minion Pattern

Inter-session task handoff for autonomous work.

​
Agent Configuration

Agents are configured in argent.json under the agents key: 
{
  "agents": {
    "defaults": {
      "model": "claude-sonnet-4-20250514",
      "maxTokens": 16384,
      "modelRouter": {
        "enabled": true
      }
    },
    "main": {
      "name": "Argent",
      "identity": "You are Argent, a personal AI assistant."
    }
  }
}

​
Lifecycle

  1. Gateway starts and loads agent configuration
  2. Channel message arrives and is routed to the agent
  3. Agent processes the message through the agent loop
  4. Response is sent back through the originating channel
  5. Memory is updated with relevant information from the interaction
The agent persists as long as the gateway is running. It wakes up on incoming messages and can also be triggered by the heartbeat system.