Quick lookup for the abbreviations and TnsAI-specific terms that appear in the rest of the documentation. Linked terms go to the page that explains the concept in depth.

Term	Definition
Action	A unit of work an agent can execute. Declared with `@ActionSpec` on a role method. See Actions.
Agent	A long-lived orchestrator. Its chat loop sends messages to the LLM and runs the tools the model calls.
BDI	Belief-Desire-Intention. The model types `Belief`, `Desire` and `Intention`, and the `CognitiveModel.bdi()` builder. `AgentBuilder` does not set one up.
Capability	A reusable contract, an interface with default-method actions, that a role can implement. See Capabilities.
Channel	An adapter for a messaging platform: Telegram, Slack, Discord, WhatsApp, email or the terminal. See Channels.
FSM	Finite state machine. The agent lifecycle (`CREATED → STARTING → RUNNING → STOPPING → STOPPED`, plus `FAILED`) is one. See State machines.
GOAP	Goal-oriented action planning: chaining actions by their preconditions and effects until a goal holds. `BackwardChainingPlanner` plans this way. See Planning.
KB / Knowledge base	A vector and BM25-indexed corpus that backs a `@Knowledge` `SEARCH` entry.
LAM	Language-Action Model, the pattern that maps LLM output to typed action calls. See LAM pattern.
MCP	Model Context Protocol. Lets agents expose tools to, or use tools from, MCP-compatible peers.
RAG	Retrieval-augmented generation.
ReAct	Reasoning and acting: the model alternates thought steps with tool calls. See ReAct.
Role	A bundle of capabilities and actions that an agent plays. See Roles.
SCOP	A separate Java multi-agent framework. See scop-framework.netlify.app. TnsAI ships an integration bridge.
SPI	Service Provider Interface, Java's `ServiceLoader` discovery pattern and the framework's main extension point. See SPI.
ToT	Tree of Thoughts, a reasoning pattern that explores several branches before it commits to one. See Tree of Thoughts.

Glossary