guided-product-demo

Warn

Audited by Gen Agent Trust Hub on Feb 22, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • EXTERNAL_DOWNLOADS (MEDIUM): The skill performs npm install for @helios-project/core, @helios-project/renderer, and @helios-project/cli. These packages are from an unvetted organization not present on the trusted sources list.
  • COMMAND_EXECUTION (MEDIUM): The skill executes npx helios render and ffmpeg via the command line. While functional for the skill's purpose, it executes code from the unvetted packages downloaded in previous steps.
  • PROMPT_INJECTION (LOW): This skill is vulnerable to Indirect Prompt Injection (Category 8).
  • Ingestion points: The skill researches external URLs provided by the user and analyzes local repository content including READMEs, marketing copy, and API endpoints (Step 3).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the analyzed data are provided.
  • Capability inventory: The agent has the ability to install packages, execute shell commands (npx, ffmpeg), and read arbitrary files in the repository.
  • Sanitization: No sanitization or validation of the content fetched from URLs or repo files is mentioned before processing.
  • CREDENTIALS_UNSAFE (SAFE): The skill requires an ELEVEN_LABS_API_KEY but follows best practices by instructing the agent not to search for .env files and instead requesting it directly from the user.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 22, 2026, 06:21 PM