1-to-1-abm-ads

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill follows established safe practices for AI agent skills. Credentials such as LINKEDIN_ACCESS_TOKEN and OPENAI_API_KEY are managed via a scripts/.env file, with a provided env.example.txt to prevent accidental inclusion of real secrets in version control. The executable components (build_campaign.py, narrow_audience.py, resolve_and_size.py, render_creatives.py) perform legitimate LinkedIn Marketing API operations and OpenAI image generation tasks consistent with the skill's stated purpose of ABM management. External downloads for dependencies (requests, python-dotenv) are from well-known official registries. No indicators of prompt injection, malicious persistence, or data exfiltration to unauthorized domains were found. The skill emphasizes human-in-the-loop safety, requiring explicit execution flags and human confirmation for activation or deletion of ad objects.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 04:38 PM
Security Audit — agent-trust-hub — 1-to-1-abm-ads