Skills
Navigate
skills/daniunai/gpt-image-2-skill/gpt-image-2/Gen Agent Trust Hub

gpt-image-2

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it interpolates raw user prompts into the instruction stream sent to the codex CLI tool's standard input.
  • Ingestion points: The user-supplied prompt argument processed in scripts/gen.js and used in scripts/image-gen.js.
  • Boundary markers: Absent; the user prompt is appended directly to an internal instruction template string.
  • Capability inventory: The skill uses child_process.spawn to execute the codex CLI, which processes these instructions.
  • Sanitization: No escaping, validation, or filtering is performed on the input prompt before interpolation.
  • [COMMAND_EXECUTION]: The skill executes the external codex CLI tool via the Node.js child_process module.
  • Evidence: The runCodex function in scripts/image-gen.js uses spawn to run the command found on the system path.
  • Security Note: While it passes arguments as an array (avoiding shell injection) and uses a --sandbox read-only flag, it relies on the integrity of the local codex installation.
  • [EXTERNAL_DOWNLOADS]: The skill documentation references the codex CLI tool from OpenAI's official GitHub repository as a required external dependency.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 07:15 AM