higgsfield-workspaces
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation and routing logic for various creative workspaces. It does not contain executable code, remote script downloads, or sensitive data access.
- [SAFE]: The metadata and instructions are consistent with the stated purpose of assisting users with platform navigation.
- [SAFE]: References to external tools like
get_presetsandexecute_presetare documented as part of the official platform schema and do not involve unauthorized command execution or network exfiltration. - [SAFE]: No prompt injection techniques, such as attempts to bypass safety filters or ignore instructions, were found.
Audit Metadata