export

Warn

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes a shell command to execute a local Python script for data persistence. \n
  • Evidence: python3 skills/ui-ux-pro-max/scripts/search.py "$QUERY" --design-system --persist -p "$PROJECT_NAME" in SKILL.md. \n
  • Risk: The use of shell variables $QUERY and $PROJECT_NAME within the command string presents a command injection risk if these values are sourced from unsanitized user input. \n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes external data from the file system. \n
  • Ingestion points: The skill reads design system definitions from design-system/MASTER.md. \n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the ingested file are provided. \n
  • Capability inventory: The skill has the capability to write multiple files to the project directory and invoke the import_html tool via the /figma skill. \n
  • Sanitization: No validation or escaping of the content extracted from MASTER.md is specified before it is interpolated into code files or HTML payloads.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 23, 2026, 08:41 AM
Security Audit — agent-trust-hub — export