abm-1-to-1
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill contains a Python script
scripts/check-output.pyused for validating generated marketing deliverables. The script performs basic length checks and keyword presence verification on local files. It does not perform network operations, access sensitive system files, or execute external code. - [INDIRECT_PROMPT_INJECTION]: The skill involves researching and processing external information about business accounts. While processing untrusted external data is a standard vector for indirect prompt injection, the skill lacks high-privilege tools, network exfiltration capabilities, or dangerous execution environments that could be exploited via this vector.
Audit Metadata