create-video-fal

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides the gen_video.py script, which serves as a command-line interface for agents to execute video generation tasks.
  • [EXTERNAL_DOWNLOADS]: The skill uses urllib.request.urlretrieve to download generated media from public FAL media URLs to a local destination specified by the user.
  • [PROMPT_INJECTION]: The skill processes potentially untrusted data through its generation payload, creating an attack surface for indirect prompt injection.
  • Ingestion points: The --payload argument in gen_video.py accepts either raw JSON strings or local file paths (using the @ prefix).
  • Boundary markers: There are no explicit delimiters or instructions in the code to isolate or ignore instructions embedded within the user-provided prompt strings.
  • Capability inventory: The skill possesses file-writing capabilities (downloading media and persisting job state in ~/.gooseworks/pending-fal-jobs/) and network access (communicating with the vendor-managed API proxy and media CDNs).
  • Sanitization: While the script performs standard JSON parsing, it does not validate or sanitize the internal content of the generation prompts before forwarding them to the external API.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 02:59 PM
Security Audit — agent-trust-hub — create-video-fal