meta-ads-expert

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The instructions do not contain attempts to bypass safety filters or override agent behavior. The persona is clearly defined and stays within the scope of managing Meta Ads.
  • [COMMAND_EXECUTION]: The skill references local helper scripts (scripts/auth_check.py) for authentication diagnostics. The provided code for these scripts is benign and follows standard practices for local development and OAuth flows.
  • [CREDENTIALS_UNSAFE]: No hardcoded secrets, tokens, or API keys were detected. The authentication workflow utilizes OAuth with a local SQLite database for secure token management, which is a recommended practice.
  • [SAFE]: The skill incorporates robust safety guardrails, including mandatory user confirmation before executing any state-changing actions such as creating or updating campaigns, or resetting the local database. It also includes strict limits on data retrieval to prevent resource exhaustion and rate-limiting issues.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 02:14 PM
Security Audit — agent-trust-hub — meta-ads-expert