photoreal-image

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The workflow ingests outsider-authored free text at runtime via the --prompt-file input to generate.sh, which reads the file contents into the JSON payload sent to image providers (jq --rawfile p "$PROMPT_FILE"), with no step that requires the agent to pre-select a trusted item.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The README instructs users to run a git clone from https://github.com/olegspiridonovdev-spec/photoreal-image-skill.git and then execute the included installer (install-everywhere.sh), which fetches remote code that would be executed locally (installation-time remote code execution risk).