backlinks-summary

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled shell script (get_backlinks_summary.sh) and embedded Python code to process data and make network requests.
  • Evidence: The script uses python3 to safely construct JSON payloads and validate response data, avoiding common shell injection vulnerabilities by using environment variables for data passing.
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to the Frevana API to retrieve backlink data.
  • Evidence: Uses curl to communicate with https://ai-factory.frevana.com. This is the intended behavior for an API-integrated skill.
  • [CREDENTIALS_UNSAFE]: The skill requires an API token (FREVANA_TOKEN) to authenticate with the remote service.
  • Evidence: The script retrieves the token from environment variables or an explicit flag and includes it in the HTTP Authorization header. It includes a specific instruction in SKILL.md and logic in the script to ensure the token is not echoed back to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 03:39 PM
Security Audit — agent-trust-hub — backlinks-summary