For the complete documentation index, see llms.txt. Markdown versions of all docs pages are available by appending .md to any docs URL.
Core Concepts
Understand the fundamental concepts and architecture of kagent.
Learn about the AI agent and kagent concepts.
The Agents page also covers prompt templates, Git-based skills, and context management. For related material, see What is kagent, the Tools Ecosystem catalog, Human-in-the-Loop configuration, and the examples, which include both the Go and Python runtimes.
kagent Architecture
Explore the high-level architecture of kagent, including its core components like the Controller, …
Agents
Dive into the concept of AI agents in kagent, their components (instructions, tools, skills), and …
Tools
Understand the different types of tools kagent can use, including built-in, MCP, and HTTP tools, and …
Agent Harness
Understand AgentHarness resources, Substrate-backed execution environments, and how kagent talks to …
Agent Substrate
Run declarative agents on Agent Substrate — a Kubernetes-native runtime that delivers fast startup, …
Agent Memory
Enable vector-backed long-term memory for agents to learn from past interactions.
MCP Apps
Render interactive MCP UI widgets (MCP Apps) inline in the kagent chat and test them from the …