agent-team-kit

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's intake and Scout role explicitly direct agents to monitor and ingest external/user-generated signals (e.g., "External signals (social mentions, community requests)" in templates/process/INTAKE.md and "External monitoring (community channels, social signals)" in templates/process/ROLES.md and HEARTBEAT.md), meaning agents are expected to read untrusted third-party content that can drive spawning, triage, and execution decisions.