team-communication-protocols

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill describes protocols for receiving and acting upon messages from other agents (via message and broadcast types), which creates a standard interface for indirect prompt injection.
  • Ingestion points: Data enters the agent's context through messages from teammates and by reading teammate metadata from ~/.claude/teams/{team-name}/config.json.
  • Boundary markers: The communication protocols do not explicitly include instructions for using delimiters or boundary markers to differentiate between instructions and untrusted data from other agents.
  • Capability inventory: The skill references capabilities to send messages (SendMessage), manage tasks (TaskUpdate), and delete team resources (TeamDelete).
  • Sanitization: No specific sanitization or validation requirements are defined for the content of messages received from teammates.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 09:58 PM
Security Audit — agent-trust-hub — team-communication-protocols