skills/borghei/claude-skills/paid-ads/Gen Agent Trust Hub

paid-ads

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of advertising guidelines and three Python scripts (ad_copy_scorer.py, audience_sizer.py, cpc_calculator.py) designed for campaign analysis. All scripts operate locally on provided input or JSON files without making network requests, executing arbitrary commands, or accessing sensitive system paths.
  • [INDIRECT_PROMPT_INJECTION]: The Python scripts ingest user-provided JSON data through file paths. However, the scripts lack dangerous capabilities such as network access or shell execution, and they only output processed data to the console. This structure prevents any malicious data from influencing the agent's environment or exfiltrating information.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 07:25 AM
Security Audit — agent-trust-hub — paid-ads