imagegen

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to run a local Python script, scripts/image_gen.py, which handles image generation and editing tasks. This script is well-documented and uses standard libraries.\n- [EXTERNAL_DOWNLOADS]: Network access is used exclusively for communication with the OpenAI Image API to generate, edit, and download images.\n- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes user-provided prompts to generate images.\n
  • Ingestion points: User-supplied strings in SKILL.md workflows and JSONL batch input files processed by scripts/image_gen.py.\n
  • Boundary markers: Absent; while prompts are augmented using a structured template, they are not strictly delimited to prevent the model from following embedded instructions.\n
  • Capability inventory: File system write operations for saving images and network operations for API requests within scripts/image_gen.py.\n
  • Sanitization: Absent; the skill performs standard CLI argument validation but does not sanitize the semantic content of image prompts.\n- [SAFE]: No malicious obfuscation, persistence mechanisms, or unauthorized data exfiltration patterns were found. The skill correctly identifies the need for an API key and provides instructions for safe local configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 03:40 AM