frevana-space-cms

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a bundled bash script (publish_file.sh) to coordinate the publishing workflow, invoking curl for API interactions and python3 for parsing JSON responses and HTML metadata. This is standard operational behavior for a CLI-based publishing tool.
  • [DATA_EXFILTRATION]: The skill transmits user-specified HTML files to api.frevana.com and vendor-provided storage endpoints over HTTPS. This transmission is required for the skill's primary function and includes appropriate authorization handling.
  • [EXTERNAL_DOWNLOADS]: The skill connects to external Frevana API endpoints to retrieve user subscription data and obtain pre-signed URLs for file uploads.
  • [PROMPT_INJECTION]: The skill ingests local HTML files and extracts titles using a Python-based parser. While this creates a surface for indirect prompt injection if the extracted text contains malicious instructions, the impact is limited by the specific data extraction logic and is an inherent risk of processing external content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 06:43 AM
Security Audit — agent-trust-hub — frevana-space-cms