Skip to content

Glossary

Glossary of Terms

  • AI Agent: An autonomous or semi-autonomous entity that performs specialized tasks within the development lifecycle.
  • Governance: A set of rules, processes, and checkpoints that ensure human control over agent actions.
  • Checkpoint: A control point where human intervention, review, or approval is required.
  • Agent Capabilities: Functions, tasks, or services that an agent can offer and negotiate.
  • Multi-agent Orchestration: Coordination of multiple agents to collaborate on complex workflows.
  • JSON-RPC 2.0: Standard messaging protocol used as the basis for communication in HugAI.
  • Initialization: The process of connecting and negotiating capabilities between agents, tools, and the system.
  • Human Approval: A decision made by a person at a critical checkpoint in the workflow.
  • Extensibility: The protocol's ability to incorporate new features, messages, or agents without breaking compatibility.
  • Traceability: An auditable record of all relevant actions, decisions, and events in the system.