Sub-Agent
A specialised agent invoked by an orchestrating parent agent to handle a bounded task — e.g. code review, data extraction, or scheduling — with its own tools, context, and success criteria. The parent decides when to delegate and how to consume the sub-agent's result. Central to Anthropic Skills and OpenAI Agents SDK patterns in 2026.
Last reviewed:
Related terms
- Multi-Agent Systems
Systems where multiple AI agents work together, each with specialized capabilities, to accomplish complex tasks. Agents may cooperate, negotiate, or compete depending on the system design.
- Skills (Anthropic)
A packaging model, introduced by Anthropic in 2026, that bundles a capability — instructions, allowed tools, reference files, and evaluation cases — into a single loadable unit that Claude can invoke as a sub-agent. Analogous to a plugin or a function, but with declarative scope and its own eval harness.
- AI Agent
An autonomous AI system that can perceive its environment, make decisions, and take actions to achieve specific goals. Agents can use tools, interact with APIs, and coordinate with other agents.
- Tool Use
The capability of AI models to invoke external tools, APIs, and systems to accomplish tasks. Modern implementations rely on structured outputs (JSON schemas or function signatures) so the model's tool calls can be parsed and executed reliably. Extends AI beyond text generation into real-world actions.
Related on this site
Framework dimensions
Free tools
Whitepapers
Comparisons
Next steps
- Take the assessment— See where you stand
Call this framework and its tools from your own agent via the Model Context Protocol (MCP) server. Works with Claude Desktop, Cursor, Zed, Continue, and the OpenAI Agents SDK.
Where this fits
Sub-Agent is part of the Agentic AI vocabulary used in the Generative AI Maturity Framework. See the full glossary for the complete set of 159 defined terms, or take the free maturity assessment to see where your organisation stands.