higgsfield-product-photoshoot

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). The package install command pipes a remote shell script from raw.githubusercontent.com into sh (curl | sh), which is a high-risk pattern for distributing malware; the other URLs are repo and image CDN links and are low risk.

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md:21-23 and 96-146 describe collecting user-provided answers (including free-form “offer / mood / hook” and “describe the product” text) and passing them as --prompt to the backend enhancer, which then submits to gpt_image_2; this is outsider-authored free text from the operating user’s conversation, not from the skill’s bundled templates.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill instructs users to run a runtime install command that fetches and pipes a remote shell script to the shell (curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh), which executes remote code and is required when the CLI is missing.