# ArgentOS Docs

## Docs

- [Accountability System](https://docs.argentos.ai/agents/accountability.md): Heartbeat contract format, task verification, scoring, penalty/reward tiers, and interval adjustments.
- [The Agent Loop](https://docs.argentos.ai/agents/agent-loop.md): How ArgentOS processes each message — receive, think, act, respond.
- [Alignment Documents](https://docs.argentos.ai/agents/alignment-docs.md): Per-agent personality and identity documents that shape agent behavior — SOUL.md, IDENTITY.md, USER.md, and more — with Git-backed versioning.
- [Agent Architecture](https://docs.argentos.ai/agents/architecture.md): How the ArgentOS agent runtime connects the gateway, Claude API, tools, and memory.
- [Argent-Native Runtime](https://docs.argentos.ai/agents/argent-runtime.md): The Argent-native agent runtime and LLM layer replacing the upstream Pi Mono dependency.
- [Consciousness Kernel](https://docs.argentos.ai/agents/consciousness-kernel.md): The persistent autonomous executive that gives ArgentOS continuous agency. Currently running in shadow mode with scheduler authority over contemplation and SIS.
- [Contemplation](https://docs.argentos.ai/agents/contemplation.md): Self-directed thinking cycles that give ArgentOS the ability to reflect, learn, and act when idle.
- [Cron & Scheduled Jobs](https://docs.argentos.ai/agents/cron-jobs.md): Scheduled job system for ArgentOS — periodic agent tasks, automated scans, nudges, and the minion pattern for autonomous background work.
- [Execution Worker](https://docs.argentos.ai/agents/execution-worker.md): Autonomous background task processor that picks up, executes, and validates tasks without human intervention.
- [Heartbeat Contracts](https://docs.argentos.ai/agents/heartbeat-contracts.md): Structured task contracts in HEARTBEAT.md — parsing, progress tracking, verification, and integration with the accountability system.
- [Agents Overview](https://docs.argentos.ai/agents/index.md): Understand how ArgentOS agents work — the AI runtime at the core of the system.
- [Knowledge Library & ACL](https://docs.argentos.ai/agents/knowledge-library.md): PostgreSQL-backed RAG system with agent-level access control, multi-format ingestion, and hybrid search.
- [The Minion Pattern](https://docs.argentos.ai/agents/minion-pattern.md): Isolated cron sessions do the grunt work. The main session does the thinking.
- [Multi-Agent](https://docs.argentos.ai/agents/multi-agent.md): Running multiple ArgentOS agents with different configurations and purposes.
- [Outbound Messaging](https://docs.argentos.ai/agents/outbound-messaging.md): Multi-channel delivery pipeline that enables agents to send messages, attachments, and polls across Telegram, Discord, Slack, WhatsApp, and more.
- [RALF + ANGEL](https://docs.argentos.ai/agents/ralf-angel.md): Response Accountability Llama Framework and the ANGEL verification loop.
- [Sessions](https://docs.argentos.ai/agents/sessions.md): How ArgentOS manages conversation sessions, context windows, and compaction.
- [Self-Improving System (SIS)](https://docs.argentos.ai/agents/sis-architecture.md): How ArgentOS learns from its own behavior — lessons, patterns, and feedback loops.
- [SpecForge Projects](https://docs.argentos.ai/agents/specforge.md): 5-stage project management conductor that guides software projects from natural language intent to approved execution.
- [System Prompt](https://docs.argentos.ai/agents/system-prompt.md): How ArgentOS assembles the system prompt from identity, soul, tools, and bootstrap sections.
- [ArgentOS Business](https://docs.argentos.ai/business/index.md): Enterprise governance, workforce orchestration, and organizational controls for ArgentOS.
- [Device Pairing & Discovery](https://docs.argentos.ai/channels/device-pairing.md): Bonjour service discovery and device registration for multi-device agent access across the local network and Tailnet.
- [Discord](https://docs.argentos.ai/channels/discord.md): Connect a Discord bot to your ArgentOS agent for server and DM interactions.
- [Google Chat](https://docs.argentos.ai/channels/google-chat.md): Connect ArgentOS to Google Workspace Chat for direct messages, group conversations, and threaded replies.
- [iMessage](https://docs.argentos.ai/channels/imessage.md): Connect ArgentOS to iMessage on macOS.
- [Channels Overview](https://docs.argentos.ai/channels/index.md): Connect ArgentOS to Telegram, Discord, WhatsApp, Slack, Signal, and iMessage.
- [Channel Pairing](https://docs.argentos.ai/channels/pairing.md): How ArgentOS links platform users to agent sessions across channels.
- [Signal](https://docs.argentos.ai/channels/signal.md): Connect ArgentOS to Signal using signal-cli.
- [Slack](https://docs.argentos.ai/channels/slack.md): Set up a Slack app to connect ArgentOS to your workspace.
- [Telegram](https://docs.argentos.ai/channels/telegram.md): Set up a Telegram bot to chat with your ArgentOS agent.
- [Channel Troubleshooting](https://docs.argentos.ai/channels/troubleshooting.md): Common issues and fixes for ArgentOS channel connections.
- [WhatsApp](https://docs.argentos.ai/channels/whatsapp.md): Connect ArgentOS to WhatsApp using the WhatsApp Web bridge.
- [argent channels](https://docs.argentos.ai/cli/channels-cmd.md): Commands for managing messaging channel connections.
- [argent gateway](https://docs.argentos.ai/cli/gateway-cmd.md): Commands for managing the ArgentOS gateway service.
- [CLI Reference](https://docs.argentos.ai/cli/index.md): All available argent CLI commands for managing your ArgentOS installation.
- [argent memory](https://docs.argentos.ai/cli/memory-cmd.md): Commands for searching and managing the agent's persistent memory.
- [argent status](https://docs.argentos.ai/cli/status-cmd.md): System health overview — check gateway, channels, memory, and service status at a glance.
- [Agent Workspace](https://docs.argentos.ai/concepts/agent-workspace.md): The workspace is your agent's home directory — alignment docs, memory logs, and skills all live here.
- [Build a Connector](https://docs.argentos.ai/connectors/build.md): Step-by-step guide to creating a new AOS connector from scratch.
- [How Connectors Work](https://docs.argentos.ai/connectors/how-it-works.md): The CLI contract, JSON output format, permission model, and discovery system behind AOS connectors.
- [Connectors Overview](https://docs.argentos.ai/connectors/index.md): AOS connectors give your agent structured access to external services — no MCP servers, no daemons, just CLI tools with JSON output.
- [Publish to Marketplace](https://docs.argentos.ai/connectors/publish.md): Package, sign, submit, and list your AOS connector on the ArgentOS Marketplace.
- [Connector Reference](https://docs.argentos.ai/connectors/reference.md): Complete list of all 61 AOS connectors with categories, resources, and status.
- [AEVP Architecture](https://docs.argentos.ai/dashboard/aevp-architecture.md): Deep dive into the Agent Expressive Visual Presence system — WebGL particle rendering, emotional color mapping, and ambient presence.
- [Dashboard Features](https://docs.argentos.ai/dashboard/features.md): Chat interface, task management, session history, model badges, and more.
- [Dashboard Overview](https://docs.argentos.ai/dashboard/index.md): The ArgentOS web dashboard — a React UI for chatting, managing tasks, and monitoring your agent.
- [Task Management UI](https://docs.argentos.ai/dashboard/tasks-ui.md): Creating tasks, tracking progress, and the play/complete workflow in the dashboard.
- [Text-to-Speech](https://docs.argentos.ai/dashboard/tts.md): ElevenLabs integration, voice selection, and TTS markers in ArgentOS.
- [API Reference](https://docs.argentos.ai/gateway/api-reference.md): WebSocket RPC protocol reference for ArgentOS gateway — connection, authentication, and all available methods.
- [Backup & Restore (Phoenix)](https://docs.argentos.ai/gateway/backup.md): Phoenix backup system for ArgentOS workspace data — local, Git, S3, and Cloudflare R2 targets with compression, encryption, and retention policies.
- [Gateway Configuration](https://docs.argentos.ai/gateway/configuration.md): Configure the gateway port, host, authentication, and service options.
- [Health Checks](https://docs.argentos.ai/gateway/health.md): Gateway health monitoring and zombie process reaping.
- [Heartbeat System](https://docs.argentos.ai/gateway/heartbeat.md): Periodic agent wake-ups for autonomous task work and proactive check-ins.
- [Gateway Overview](https://docs.argentos.ai/gateway/index.md): The always-on WebSocket control plane that manages agents, channels, and sessions.
- [Remote Access](https://docs.argentos.ai/gateway/remote-access.md): Access your ArgentOS gateway remotely via Tailscale, SSH tunnels, or reverse proxies.
- [Gateway Security](https://docs.argentos.ai/gateway/security-ops.md): Sandboxing, tool policies, and exec approval workflows.
- [FAQ](https://docs.argentos.ai/help/faq.md): Frequently asked questions about ArgentOS.
- [Help & Support](https://docs.argentos.ai/help/index.md): Get help with ArgentOS — support channels, community, and resources.
- [Troubleshooting](https://docs.argentos.ai/help/troubleshooting.md): Common ArgentOS issues and how to fix them.
- [ArgentOS Documentation](https://docs.argentos.ai/index.md): The operating system for personal AI — self-hosted, always-on, connected everywhere.
- [Developer Branch Installs](https://docs.argentos.ai/install/developer-branch.md): Install ArgentOS from the private develop lane to test release candidates before promotion to main.
- [Docker](https://docs.argentos.ai/install/docker.md): Run ArgentOS in a Docker container.
- [Installation](https://docs.argentos.ai/install/index.md): Install ArgentOS on macOS, Linux, or via Docker.
- [Node.js Requirements](https://docs.argentos.ai/install/node-requirements.md): Node.js version requirements and native module compatibility.
- [Uninstall](https://docs.argentos.ai/install/uninstall.md): Remove ArgentOS from your system.
- [Updating](https://docs.argentos.ai/install/updating.md): Keep ArgentOS up to date.
- [Intent & Safety](https://docs.argentos.ai/intents/index.md): Control what your agent can and cannot do — from simple safety rules to full enterprise governance.
- [Industry Packs](https://docs.argentos.ai/intents/industry-packs.md): Pre-built intent configurations tuned for specific business verticals.
- [Safety Rules](https://docs.argentos.ai/intents/safety-rules.md): Set basic boundaries for your agent — what it should never do and what needs your approval.
- [Simulation Gates](https://docs.argentos.ai/intents/simulation-gates.md): Test agent behavior against quality benchmarks before allowing deployment.
- [License Activation](https://docs.argentos.ai/licensing/activation.md): Activate your ArgentOS license — key entry, hardware fingerprint, and verification.
- [Enterprise Features](https://docs.argentos.ai/licensing/enterprise.md): Organizations, API key management, instance management, and secret sync for teams.
- [Licensing Overview](https://docs.argentos.ai/licensing/index.md): How ArgentOS licensing works — plans, features, and activation.
- [Atera MSP Integration](https://docs.argentos.ai/marketplace/atera.md): Connect ArgentOS to your Atera RMM/PSA platform — tickets, devices, alerts, and reports through natural language.
- [Browsing & Installing](https://docs.argentos.ai/marketplace/browsing.md): Search, evaluate, and install packages from the ArgentOS marketplace.
- [Marketplace Overview](https://docs.argentos.ai/marketplace/index.md): Browse and install skills, plugins, avatars, and templates from the ArgentOS marketplace.
- [Publishing Packages](https://docs.argentos.ai/marketplace/publishing.md): Create and share your own skills, plugins, avatars, and templates.
- [Memory Architecture Lanes](https://docs.argentos.ai/memory/architecture-lanes.md): MemU, RAG Library, and SIS working together for continuity.
- [Consciousness Memory (Observation Layer)](https://docs.argentos.ai/memory/consciousness-memory.md): Experimental synthesis layer bridging raw memory substrates and the Consciousness Kernel through durable knowledge observations.
- [Memory Health Runbook](https://docs.argentos.ai/memory/health-runbook.md): Daily operator checks and incident triage for MemU, contemplation, SIS, and RAG.
- [Identity Fields](https://docs.argentos.ai/memory/identity-fields.md): Rich metadata in MemU — emotional context, entities, relationships, and personal attributes.
- [Memory Overview](https://docs.argentos.ai/memory/index.md): How ArgentOS remembers — MemU is the agent's persistent brain.
- [Lossless Context Management (LCM)](https://docs.argentos.ai/memory/lcm.md): DAG-based context compression that preserves every message with full recall via grep, describe, and expand tools.
- [MemU Architecture](https://docs.argentos.ai/memory/memu.md): Inside MemU — PostgreSQL + pgvector, hybrid search, auto-capture, significance scoring, and embeddings.
- [PG+Redis Architecture](https://docs.argentos.ai/memory/pg-redis.md): PostgreSQL and Redis migration architecture — three-phase cutover, StorageAdapter interface, schema design, and configuration.
- [Memory Profile Collapse](https://docs.argentos.ai/memory/profile-collapse.md): Automated deduplication of operational profile snapshots in the MemU memory system.
- [Semantic Search](https://docs.argentos.ai/memory/semantic-search.md): How MemU searches memories using FTS5 queries, embedding similarity, and reranking.
- [SIS Reliability](https://docs.argentos.ai/memory/sis-reliability.md): How SIS consolidation works, fails, and should be monitored.
- [Auth Profiles](https://docs.argentos.ai/models/auth-profiles.md): Manage multiple API keys and subscriptions with automatic failover.
- [Model Failover](https://docs.argentos.ai/models/failover.md): Automatic failover with exponential backoff, cooldowns, and profile rotation.
- [Model System Overview](https://docs.argentos.ai/models/index.md): How ArgentOS routes tasks to the right model tier for optimal cost and performance.
- [Provider Registry](https://docs.argentos.ai/models/provider-registry.md): Dynamic model provider catalog with 15+ providers, persistent customization, and automatic seed updates for ArgentOS's multi-provider LLM infrastructure.
- [Providers](https://docs.argentos.ai/models/providers.md): Supported model providers — Anthropic, Ollama, MiniMax, Z.AI, and OpenRouter.
- [Router Architecture](https://docs.argentos.ai/models/router-architecture.md): Deep dive into the model router — complexity scoring, tier routing, background lanes, and cross-provider fallback.
- [Authentication](https://docs.argentos.ai/security/auth.md): API keys, setup tokens, auth profiles, and gateway authentication.
- [Encrypted Secrets](https://docs.argentos.ai/security/encrypted-secrets.md): AES-256-GCM encryption for API keys and credentials at rest, with OS keychain integration and PostgreSQL secret store.
- [Exec Approvals](https://docs.argentos.ai/security/exec-approvals.md): Command-level gating that controls which shell commands agents can execute, with allowlists, IPC approval daemons, and per-agent policies.
- [Security Overview](https://docs.argentos.ai/security/index.md): How ArgentOS protects your system — sandboxing, authentication, and secret management.
- [Sandboxing](https://docs.argentos.ai/security/sandboxing.md): Control what your agent can do with sandboxing modes and tool policies.
- [Secret Management](https://docs.argentos.ai/security/secrets.md): Secure storage and distribution of API keys, tokens, and credentials.
- [Configuration Reference](https://docs.argentos.ai/start/config-reference.md): Complete reference for argent.json — all configuration sections, fields, defaults, and environment variable overrides.
- [Configuration](https://docs.argentos.ai/start/configuration.md): Configure ArgentOS via argent.json and environment variables.
- [Getting Started](https://docs.argentos.ai/start/getting-started.md): Install ArgentOS and talk to your personal AI in under 5 minutes.
- [Learning Path](https://docs.argentos.ai/start/learning-path.md): Find the right docs for your experience level. From first install to enterprise governance.
- [Legacy Migration Reference](https://docs.argentos.ai/start/migration.md): Historical reference for transitioning from a legacy OpenClaw installation to ArgentOS.
- [Onboarding](https://docs.argentos.ai/start/onboarding.md): Step-by-step guide to setting up your ArgentOS agent.
- [Uninstall ArgentOS](https://docs.argentos.ai/start/uninstall.md): How to remove ArgentOS from your machine. Covers macOS, with Linux and Windows guides coming soon.
- [Creating Tasks](https://docs.argentos.ai/tasks/creating-tasks.md): How to create and manage tasks via chat, dashboard, CLI, and agent tools.
- [Task System Overview](https://docs.argentos.ai/tasks/index.md): ArgentOS task system — a priority queue with accountability for your AI agent.
- [Projects](https://docs.argentos.ai/tasks/projects.md): Multi-step task grouping — project lifecycle, agent tools, and dashboard integration.
- [Task Lifecycle](https://docs.argentos.ai/tasks/task-flow.md): The full task flow — pending to in_progress to completed, with dashboard integration.
- [Browser Automation](https://docs.argentos.ai/tools/browser.md): The browser tool — navigate, interact with, and extract content from web pages.
- [Building Plugins](https://docs.argentos.ai/tools/building-plugins.md): End-to-end guide to building ArgentOS plugins — tools, hooks, service keys, and config persistence.
- [Command Execution](https://docs.argentos.ai/tools/exec.md): The exec tool — run bash commands, scripts, and system operations from your agent.
- [Tools Overview](https://docs.argentos.ai/tools/index.md): The 50+ tools available to ArgentOS agents — file operations, web research, memory, tasks, knowledge, communication, and more.
- [MCP Integration](https://docs.argentos.ai/tools/mcp.md): Connect to MCP servers to extend ArgentOS with external tools and data sources. Configure stdio and HTTP servers with filtering and security controls.
- [MCP Platform Server](https://docs.argentos.ai/tools/mcp-server.md): Expose ArgentOS as an MCP server so Claude Desktop, Cursor, Claude Code, and any MCP client can operate your agent system remotely — memory, tasks, projects, and direct agent conversation.
- [Plugins](https://docs.argentos.ai/tools/plugins.md): The ArgentOS plugin system — extend agent behavior with argent.plugin.json manifests.
- [Skills, Plugins & Connectors](https://docs.argentos.ai/tools/skills.md): Extend your agent with marketplace packages. Skills teach, plugins add runtime capabilities, connectors wire to external systems. Install with one command.
- [Slash Commands](https://docs.argentos.ai/tools/slash-commands.md): Quick actions you can trigger from any channel with slash command syntax.
- [TOON Encoding](https://docs.argentos.ai/tools/toon-encoding.md): How ArgentOS uses Token-Oriented Object Notation to save 40-50% on LLM prompt tokens.
- [Web Search & Fetch](https://docs.argentos.ai/tools/web.md): Give your agent the ability to search the web and fetch content from URLs.
- [Control UI (Dashboard)](https://docs.argentos.ai/web/control-ui.md): The web-based dashboard for managing your ArgentOS agent — chat, tasks, configuration, and more.

## OpenAPI Specs

- [openapi](https://docs.argentos.ai/api-reference/openapi.json)

## Optional

- [GitHub](https://github.com/ArgentAIOS)
- [Marketplace](https://marketplace.argentos.ai)
- [Community](https://discord.gg/argentos)