guided-explainer-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 several Node.js packages (@helios-project/core, @helios-project/renderer, @helios-project/cli) during execution. These packages originate from an external organization not included in the trusted sources list.
  • [COMMAND_EXECUTION] (LOW): The skill executes external CLI tools including ffmpeg (for audio analysis) and npx helios render (for video production). While these are aligned with the skill's stated purpose, they represent execution of code from unverified external packages.
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface: The skill fetches content from external URLs (Step 3) to research brand identity. Evidence Chain: 1. Ingestion point: Step 3 fetches user-provided URLs. 2. Boundary markers: Absent; fetched content is processed directly. 3. Capability inventory: Skill can install packages, write files, and execute CLI commands. 4. Sanitization: Absent; no validation or escaping of fetched content is performed before analysis.
  • [CREDENTIALS_UNSAFE] (SAFE): The skill correctly handles sensitive credentials by prompting the user to provide the ELEVEN_LABS_API_KEY in the environment and explicitly instructs the agent not to attempt to read local .env files.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 22, 2026, 06:21 PM