character-animation-qa

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from project files and web content, creating a potential surface for indirect instructions to influence agent behavior.
  • Ingestion points: Character animation schema files, static project assets, and the content of the previewUrl provided during the QA process.
  • Boundary markers: The skill does not provide delimiters or instructions to the agent to treat external content as untrusted or to ignore embedded commands.
  • Capability inventory: The agent is instructed to use Playwright for browser automation and FFmpeg/ffprobe for media analysis, providing significant interaction capabilities with the local environment and network.
  • Sanitization: No sanitization or validation logic is specified for the data ingested during the review steps.
  • [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands using the ffmpeg and ffprobe utilities to validate media duration, resolution, and metadata.
  • [DYNAMIC_EXECUTION]: The skill provides a TypeScript code template for using Playwright to launch a browser, navigate to external URLs, and capture screenshots.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 08:42 AM