image-gen-blockeden-openai
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructions establish an indirect prompt injection surface where untrusted user input is interpolated into shell commands.\n
- Ingestion points: User-supplied descriptions are used as the
<prompt>argument for thescripts/generate_image.pyscript as described in SKILL.md.\n - Boundary markers: The suggested command template uses double quotes (
"<prompt>"), which is an inadequate boundary marker for preventing shell injection if the input contains escaped characters.\n - Capability inventory: The
scripts/generate_image.pyscript has capabilities to perform network requests (viaurllib.request) and write files to the local system (viaopen(filepath, "wb")).\n - Sanitization: No sanitization, escaping, or validation logic is present in the skill's Python script or instructions to handle potentially malicious input strings.
Audit Metadata