senior-frontend

Warn

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs shell operations to verify the environment (node, python3) and relies on the execution of local Python scripts (scripts/frontend_scaffolder.py, scripts/component_generator.py, scripts/bundle_analyzer.py) for its core functionality.
  • [PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection and command injection by interpolating user-controlled data into shell command arguments.
  • Ingestion points: User-provided names for projects and components entered in response to SKILL.md instructions.
  • Boundary markers: No delimiters or instructions to ignore embedded commands are present in the script execution workflows.
  • Capability inventory: The skill possesses the ability to execute shell scripts and invoke package managers (npm, npx).
  • Sanitization: There is no evidence of input validation, character escaping, or sanitization to prevent malicious shell sequences in user-provided strings.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 5, 2026, 08:06 PM
Security Audit — agent-trust-hub — senior-frontend