Skills
skills/nickcrew/claude-ctx-plugin/skill-installer/Gen Agent Trust Hub

skill-installer

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches skill configurations and repository archives from GitHub's official API and codeload service.
  • Evidence: Found in scripts/github_utils.py and scripts/install-skill-from-github.py using urllib.request.
  • The default repository is openai/skills, which is a verified trusted organization.
  • [COMMAND_EXECUTION]: Utilizes the git command-line tool via the subprocess module to manage repository sparse checkouts.
  • Evidence: Implemented in scripts/install-skill-from-github.py using subprocess.run with list-style arguments to prevent shell injection.
  • The skill restricts git operations to repositories and paths specified by the user or defaults.
  • [REMOTE_CODE_EXECUTION]: Facilitates the download and placement of external skill scripts into the local installation directory.
  • Evidence: The core function of scripts/install-skill-from-github.py is to extract and copy repository content into $CODEX_HOME/skills.
  • Mitigation: The _safe_extract_zip function explicitly checks extracted file paths to ensure they stay within the target destination, effectively preventing directory traversal (zip-slip) attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 08:05 PM