skill-installer

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads content from github.com and codeload.github.com. These are well-known, trusted services for software distribution. The implementation correctly handles authentication via GITHUB_TOKEN when provided.
  • [COMMAND_EXECUTION]: The script scripts/install-skill-from-github.py uses subprocess.run to execute git commands. This is a functional requirement for the skill's purpose (installing code from repositories). The commands are constructed using argument lists rather than shell strings, which prevents common shell injection vulnerabilities.
  • [SAFE_PRACTICE]: The skill implements path traversal protection in _safe_extract_zip by verifying that all extracted files remain within the intended destination directory. It also validates that skill names are single path segments to prevent arbitrary file system writes.
  • [REMOTE_CODE_EXECUTION]: While the skill installs code that will later be executed by the agent platform, it does not execute the downloaded code during the installation process itself. It serves as a package manager utility.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 12:40 PM
Security Audit — agent-trust-hub — skill-installer