# Docs

## PROTOCOL

- [Architecture](/docs/architecture): High-level architecture of the Agentic OS Protocol
- [Introduction](/docs): An overview of the Agentic OS Protocol
- [Motivation](/docs/motivation): Why the Agentic OS Protocol exists and what problems it solves
- [MCP Servers](/docs/actions/mcp-servers): Model Context Protocol server integration
- [Operations](/docs/actions/ops): Core operations and actions for agent execution
- [Tools](/docs/actions/tools): Tool management and execution for AI agents
- [Schema](/docs/apps/schema): Application schema definitions and data structures
- [Agent Loop](/docs/concepts/agent-loop): The core execution pattern for AI agents
- [Agentic OS](/docs/concepts/agentic-os): The architectural metaphor that defines how AI systems manage agent resources
- [Agent Lifecycle](/docs/concepts/lifecycle): System-level agent management workflow
- [Workflows Taxonomy](/docs/concepts/workflows-taxonomy): Understanding the six categories of workflow patterns
- [Audit](/docs/checks/audit): Agent behavior monitoring and compliance auditing
- [Fallback](/docs/checks/fallback): Error handling and recovery mechanisms
- [Judge](/docs/checks/judge): Agent performance and output quality evaluation
- [Rules](/docs/checks/rules): Behavioral constraints and guidelines for agents
- [Screenshot](/docs/checks/screenshot): Visual validation and monitoring capabilities
- [Apps Context](/docs/context/apps): Application-specific context and data management
- [Documents](/docs/context/documents): Document processing and management
- [Embeddings](/docs/context/embeddings): Vector embeddings generation and management
- [Context](/docs/context): Application-specific context and data management
- [Memory](/docs/context/memory): Persistent memory and knowledge storage
- [Persistence](/docs/context/persistence): Data persistence and retrieval mechanisms
- [System Context](/docs/context/system): System-level context and configuration
- [Cancel](/docs/runs/cancel): Task and run cancellation mechanisms
- [Human-in-the-Loop](/docs/runs/human-in-the-loop): Human oversight and intervention mechanisms
- [Retries](/docs/runs/retries): Retry mechanisms for failed operations
- [Timeout](/docs/runs/timeout): Timeout management for agent operations
- [Executor](/docs/skills/executor): Execute specific tasks and operations
- [Orchestrator](/docs/skills/orchestrator): Coordinate multiple agents and manage task distribution
- [Planner](/docs/skills/planner): Break down complex objectives into actionable tasks
- [Environment](/docs/system/env): Configuration and environment management
- [Filesystem](/docs/system/fs): File system operations and management
- [Introduction](/docs/system): Infrastructure services that provide system-level intelligence for agent orchestration
- [Installer](/docs/system/installer): System installation and setup
- [MCP Client](/docs/system/mcp-client): Model Context Protocol client implementation
- [Preferences](/docs/system/preferences): Agent preferences and user settings
- [Registry](/docs/system/registry): Agent registration and discovery system
- [Settings](/docs/system/settings): System and agent settings management
- [Evaluator Optimizer](/docs/workflows/evaluator-optimizer): Workflow evaluation and optimization
- [Orchestrator Worker](/docs/workflows/orchestrator-worker): Worker pattern for orchestrator workflows
- [Parallelization](/docs/workflows/parallelization): Parallel task execution and workflow parallelization
- [Routing](/docs/workflows/routing): Task routing and workflow routing strategies