guided-promo-video

Warn

Audited by Gen Agent Trust Hub on Feb 22, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • EXTERNAL_DOWNLOADS (MEDIUM): The skill installs @helios-project/core, @helios-project/renderer, and @helios-project/cli using npm. These packages do not belong to the trusted organizations list, posing a risk of malicious dependency execution if the registry or account is compromised.
  • PROMPT_INJECTION (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8) due to its design:
  • Ingestion points: In STEP 3, the skill fetches content from user-provided marketing URLs and reads various repository files (CSS, Tailwind configs, and marketing copy).
  • Boundary markers: Absent. There are no instructions to the agent to ignore or delimit instructions found within the fetched brand data.
  • Capability inventory: The skill can install packages, write local files (composition.html), and execute shell commands (npx helios, ffmpeg).
  • Sanitization: Absent. The agent is instructed to "focus on visible brand elements" and "align all creative decisions" with the extracted data, which could include hidden malicious instructions.
  • COMMAND_EXECUTION (LOW): The skill executes ffmpeg and npx helios. While these are functional requirements for video generation, they are triggered following the processing of potentially untrusted external data.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 22, 2026, 06:21 PM