github-cli-health-check

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes GitHub CLI commands (gh auth status, gh repo list, gh api rate_limit) to assess the health of the environment. In Cowork environments, it routes these commands to the host machine via the Desktop Commander tool (mcp__Desktop_Commander__start_process) at a hardcoded path (/opt/homebrew/bin/gh).- [DATA_EXPOSURE]: The skill reads repository metadata and authentication status to generate its report. It includes safety rules to ensure it never mutates GitHub state, never echos raw tokens, and saves reports outside of git repositories to prevent accidental leaks.- [PROMPT_INJECTION]: The skill includes an indirect prompt injection surface as it processes output from external commands (gh repo list). Ingestion points: Command output from GitHub repository listings. Boundary markers: None. Capability inventory: Host process execution via Desktop Commander and file system writes for report generation. Sanitization: No explicit sanitization of command output is described.- [SAFE]: The skill references an official GitHub issue from the Anthropics repository to document known network limitations in specific environments, facilitating transparent troubleshooting for users.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 01:39 PM
Security Audit — agent-trust-hub — github-cli-health-check