web-shader-extractor

Fail

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill relies on external code fetched from an untrusted GitHub repository (github.com/lixiaolin94/skills) that is not affiliated with the author or a known trusted organization.
  • [REMOTE_CODE_EXECUTION]: Instructions provide a command (npx skills add) that downloads and executes code from a remote URL, creating a significant supply chain risk.
  • [COMMAND_EXECUTION]: The core functionality involves generating a 'standalone runnable' from web resources, which implies the creation and execution of dynamic code based on external, untrusted inputs.
  • [PROMPT_INJECTION]: The skill processes external webpage data to generate project code, creating an attack surface for indirect prompt injection.
  • Ingestion points: Live webpages targeted for shader/WebGL extraction (SKILL.md).
  • Boundary markers: No explicit sanitization or delimiters for external data are defined in the instructions.
  • Capability inventory: File writing and generation of executable JS projects.
  • Sanitization: No evidence of validation or escaping for extracted web resources.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jul 18, 2026, 08:43 AM
Security Audit — agent-trust-hub — web-shader-extractor