higgsfield-product-photoshoot
Fail
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructions direct the agent to download a script from a remote URL (
https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh) and pipe it directly into the shell (| sh) for installation. This is a critical execution pattern that allows for arbitrary code to run on the host system without prior inspection. - [EXTERNAL_DOWNLOADS]: The skill relies on fetching an external installation script from a GitHub repository that is not associated with the skill author, introducing a supply chain risk.
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to execute thehiggsfieldCLI. It interpolates user-controlled data into the shell command via the--promptargument. While the instructions suggest the backend assembles the prompt, the agent is still responsible for constructing and executing the command string, creating a surface for potential command injection if input is not properly sanitized.
Recommendations
- HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata