EchoGuard: An Agentic Framework with Knowledge-Graph Memory for Detecting Manipulative Communication in Longitudinal Dialogue

TL;DR

EchoGuard is an agentic AI framework that uses a knowledge graph to track and detect manipulative communication patterns in long-term dialogues, helping users recognize subtle manipulation tactics.

Contribution

It introduces a novel agentic framework with a knowledge graph memory for detecting manipulative communication in longitudinal dialogues, combining structured logging, complex graph queries, and targeted prompts.

Findings

Uses a knowledge graph for structured memory of interactions

Detects six psychologically-grounded manipulation patterns

Generates targeted Socratic prompts for user self-discovery

Abstract

Manipulative communication, such as gaslighting, guilt-tripping, and emotional coercion, is often difficult for individuals to recognize. Existing agentic AI systems lack the structured, longitudinal memory to track these subtle, context-dependent tactics, often failing due to limited context windows and catastrophic forgetting. We introduce EchoGuard, an agentic AI framework that addresses this gap by using a Knowledge Graph (KG) as the agent's core episodic and semantic memory. EchoGuard employs a structured Log-Analyze-Reflect loop: (1) users log interactions, which the agent structures as nodes and edges in a personal, episodic KG (capturing events, emotions, and speakers); (2) the system executes complex graph queries to detect six psychologically-grounded manipulation patterns (stored as a semantic KG); and (3) an LLM generates targeted Socratic prompts grounded by the subgraph of detected patterns, guiding users toward self-discovery. This framework demonstrates how the interplay between agentic architectures and Knowledge Graphs can empower individuals in recognizing manipulative communication while maintaining personal autonomy and safety. We present the theoretical foundation, framework design, a comprehensive evaluation strategy, and a vision to validate this approach.