Awareness Module

Overview

The Awareness module is responsible for understanding the context of each interaction. It analyzes incoming messages to extract entities, detect intent, identify topics, and assess emotional tone. This extracted context is shared with other modules during the Context Building phase of the agent pipeline.

Context Extraction

When a message arrives, the Awareness module performs several analyses in parallel. Entity extraction identifies people, places, organizations, and other named entities mentioned in the input. Intent detection classifies what the user is trying to accomplish. Topic modeling identifies the subject areas under discussion. Together, these analyses create a rich context object that informs the agent's reasoning.

Integration with Other Modules

The Awareness module's output drives behavior across the entire platform. The Memory module uses extracted entities and topics to query relevant memories. The Social Network module uses identified entities to look up relationship context. The Event Memory module uses the awareness context to determine whether an interaction qualifies as a significant episode worth storing.

MCP Tools

The Awareness module exposes tools for explicit context analysis. The LLM can invoke awareness_analyze to perform deep analysis on a piece of text, or awareness_entities to extract a structured list of entities from content. These tools are particularly useful when the agent needs to process documents or external data during a conversation.