agent/agent_models library
Classes
- AgentContext
- A window of information injected into every LLM prompt.
- AgentDecision
- The action decision produced by the LLM Core after reasoning.
- AgentGoal
- The top-level objective the agent is working toward.
- AgentMemoryEntry
- A long-term memory entry specific to the agent framework.
- AgentObservation
- A piece of information the agent has gathered from the environment (tool results, user inputs, sensor data, etc.).
- AgentRunResult
- Summary produced when the agent loop terminates.
- AgentTask
- A single concrete step produced by the PlanningEngine.
- TaskResult
- The outcome of executing an AgentTask.
Enums
- AgentDecisionType
- What kind of action the agent decided to take.
- AgentMemoryType
- Category of an AgentMemoryEntry.
- TaskStatus
- Execution status for a AgentTask.