Skills
skills/qwwiwi/agentos-skills-public/higgsfield-soul-id/Gen Agent Trust Hub

higgsfield-soul-id

Fail

Audited by Gen Agent Trust Hub on Jul 19, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructs the agent to install the Higgsfield CLI tool by downloading a script and piping it directly into the shell (curl | sh). This pattern executes external code without integrity verification.
  • Evidence: SKILL.md contains the command curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh.
  • [COMMAND_EXECUTION]: The skill constructs shell commands by interpolating user-provided inputs, such as the character name and image paths, directly into the CLI arguments. This lacks explicit sanitization and creates a surface for shell command injection.
  • Evidence: In SKILL.md, the command higgsfield soul-id create --name "<name>" --soul-2 --image <image_path> uses the user-supplied <name> and <image_path> variables.
  • [EXTERNAL_DOWNLOADS]: The skill downloads and installs software from a third-party GitHub repository (higgsfield-ai/cli) that is not recognized as a trusted source in the provided configuration.
  • Evidence: SKILL.md references the URL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh for bootstrapping.
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
Risk Level
HIGH
Analyzed
Jul 19, 2026, 05:16 PM
Security Audit — agent-trust-hub — higgsfield-soul-id