command-creator

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • COMMAND_EXECUTION (LOW): The skill utilizes the Bash tool to perform repository checks and create local directories for command storage.
  • REMOTE_CODE_EXECUTION (LOW): It generates markdown files that the agent executes as shell commands, which is the intended use case for developer automation.
  • PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection as it transforms untrusted user input into persistent agent instructions. Ingestion points: User-provided command names, descriptions, and workflow steps in SKILL.md. Boundary markers: Absent in the resulting command file templates. Capability inventory: Access to Bash for command execution and Write for filesystem modifications. Sanitization: No technical sanitization of user input is performed beyond natural language instructions.
  • PERSISTENCE (LOW): The skill writes executable content to the user's global and project-level command directories to enable reusable workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 11:27 PM