Command Development

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional, providing documentation for developers to build slash commands for Claude Code.
  • [COMMAND_EXECUTION]: The documentation describes the platform's native support for bash execution via the !command`` syntax. All provided examples (e.g., git diff, npm test, gh pr view) are standard development workflows used as educational templates.
  • [EXTERNAL_DOWNLOADS]: Mentions external tools and package managers like npm, git, and gh (GitHub CLI) within the context of developer commands. No unauthorized or suspicious downloads are initiated by the skill itself.
  • [DATA_EXFILTRATION]: While the skill explains how to reference files using the @ syntax, it is presented as a feature for Claude to gather context during a development session, which is the intended purpose of the platform.
  • [PROMPT_INJECTION]: The skill provides templates that interpolate user arguments ($ARGUMENTS, $1). It correctly advises developers to validate these inputs and check for file existence, which is a security best practice for command development.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 11:26 PM
Security Audit — agent-trust-hub — Command Development