marketing-skills-collection

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a functional alias designed to maintain backward compatibility for marketing automation tasks. It does not contain executable code or malicious instructions.
  • [COMMAND_EXECUTION]: No suspicious shell commands, privilege escalation (sudo), or unauthorized file system operations were identified. The instructions focus on text-based routing and brief generation.
  • [DATA_EXFILTRATION]: There are no hardcoded credentials, sensitive file path accesses (e.g., .ssh, .env), or instructions to exfiltrate data to untrusted external domains.
  • [REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts. It only references local markdown files and peer skill definitions within the same directory structure.
  • [PROMPT_INJECTION]: No evidence of system prompt extraction, safety filter bypasses, or instructions to ignore previous rules was found in the instructions or metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 07:06 AM
Security Audit — agent-trust-hub — marketing-skills-collection