multi-agent-orchestration

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a documentation guide for multi-agent workflows within the Claude Code environment. It outlines the usage of orchestration tools like Agent, SendMessage, and TeamCreate without executing any malicious commands or accessing sensitive data.
  • [PROMPT_INJECTION]: The skill includes defensive best practices to mitigate indirect prompt injection, specifically recommending that a coordinator agent synthesize findings from research workers before delegating implementation tasks. This digest step prevents the direct flow of potentially untrusted data from the web or external files into implementation prompts.
  • [DATA_EXFILTRATION]: No patterns of data exfiltration or unauthorized credential access were detected. The mentioned WebSearch and WebFetch tools are standard capabilities for information gathering agents within this platform.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 02:42 PM
Security Audit — agent-trust-hub — multi-agent-orchestration