gpt-image-2
Pass
Audited by Gen Agent Trust Hub on Apr 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches a community prompt library from a public GitHub repository and references third-party services for image uploads.
- [COMMAND_EXECUTION]: The provided Bash, Python, and Node.js recipes include operations for creating local directories and writing image files to the local file system.
- [PROMPT_INJECTION]: The skill suggests fetching a remote README file from a third-party GitHub repository to provide the agent with prompt examples, which introduces a surface for indirect prompt injection if the remote source is modified.
Audit Metadata