skills/shipshitdev/skills/image/Gen Agent Trust Hub

image

Pass

Audited by Gen Agent Trust Hub on Oct 31, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill orchestrates the codex CLI to perform image generation tasks. It utilizes temporary files and quoted heredocs (<<'PROMPT') to handle prompt data, which effectively prevents shell variable expansion and command injection vulnerabilities during execution.
  • [COMMAND_EXECUTION]: Employs standard system utilities such as grep, awk, and the macOS sips tool for log parsing and image post-processing (resizing and format conversion).
  • [DATA_EXFILTRATION]: Accesses the local directory ~/.codex/sessions/ to recover base64-encoded image data from session rollout files. This access is a functional requirement for retrieving images generated by the headless codex exec command and is scoped to the specific session ID generated during the skill's execution. No unauthorized data access or external exfiltration patterns were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 31, 2026, 01:13 PM