Trust exploitation between agents

https://taxonomy.eticas.ai/risk/trust-exploitation-between-agents

Maturity: established

A compromised or malicious agent manipulates other agents in a pipeline by exploiting implicit trust in inter-agent communication, passing false instructions, forged outputs, or poisoned context to downstream agents.

System type: Large language models (LLM)
Lifecycle stages: Post Processing

Mappings to external frameworks

Standards & frameworks

Framework Reference
OWASP Top 10 for Agentic Applications 2026 ASI07 Insecure Inter-Agent Communication

Taxonomies & vocabularies

Framework Reference
MIT AI Risk Repository Multi-agent risks