higgsfield-brandkit
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads a CLI installer from the author's official GitHub repository (higgsfield-ai/cli). This is a standard installation method for the vendor's own tooling and is performed only after user permission is obtained.
- [COMMAND_EXECUTION]: The skill uses the Python
subprocessmodule to call system utilities including ImageMagick, LibreOffice, and Poppler for image and document processing. These calls are constructed using list-based arguments, which prevents command injection by ensuring inputs are not interpreted as shell commands. - [DATA_EXFILTRATION]: The
brandkit.pyscript contains a network fetching utility that validates hostnames before making requests. It verifies that the target IP address is global and public, which mitigates Server-Side Request Forgery (SSRF) risks and prevents unauthorized access to the user's local network resources. - [REMOTE_CODE_EXECUTION]: The skill implements a strict SVG validation engine that scans for and rejects potentially malicious content such as embedded
<script>tags, event handlers (e.g.,onclick), and external CSS imports. This ensures that user-supplied or generated visual assets do not trigger unintended code execution during processing.
Audit Metadata