puzzle-image-generator

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The script scripts/generate_puzzle_image.js fetches image data from Google's Gemini API (generativelanguage.googleapis.com). This is a well-known service and the activity is essential to the skill's purpose.\n- [COMMAND_EXECUTION]: The script scripts/generate_puzzle_image.js performs file system operations, including directory creation and file writing. It uses user-supplied arguments to determine the storage location.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external strings for command-line arguments.\n
  • Ingestion points: The --prompt, --out, and --name arguments in scripts/generate_puzzle_image.js are populated by the agent based on user requests.\n
  • Boundary markers: None; the script does not use delimiters to isolate these inputs.\n
  • Capability inventory: The script has file write capabilities (fs.writeFileSync) and network access (fetch) in scripts/generate_puzzle_image.js.\n
  • Sanitization: There is no validation or sanitization of the --name or --out arguments, which could lead to path traversal (e.g., writing to sensitive system directories) if malicious inputs are provided to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 04:19 PM
Security Audit — agent-trust-hub — puzzle-image-generator