skills/openai/skills/skill-installer/Gen Agent Trust Hub

skill-installer

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • Network Operations: The skill initiates connections to api.github.com and codeload.github.com to list and download skill contents. Interacting with remote servers is a security consideration as it involves processing external data, but this is used here to facilitate legitimate updates and installations from trusted sources.\n- Command Execution: The skill executes the git binary to perform repository operations. Running system commands with parameters derived from user input (like repository paths) is a pattern that requires careful validation; in this case, the skill uses these commands to provide standard installation workflows.\n- File System Interaction: It writes files to the user's home directory under ~/.codex/skills. Managing local files is a core capability that warrants attention to ensure only intended locations are modified, and the skill includes logic to avoid unexpected overwrites.\n- Path Traversal Protection: The code includes a check in the _safe_extract_zip function to ensure files cannot be written outside the target directory. This is an important security control that mitigates risks associated with processing potentially untrusted ZIP archives.\n- Credential Usage: The skill can utilize environment variables like GITHUB_TOKEN for authentication. While handling credentials requires care, this implementation follows standard practices for authorizing requests to the GitHub API.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 07:00 AM