ccg

Warn

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill constructs shell commands by interpolating AI-generated prompts into double-quoted strings (e.g., omc ask codex "<codex prompt>"). This pattern is vulnerable to shell injection if the prompt contains shell metacharacters like backticks, semicolons, or unmatched double quotes, which could allow arbitrary command execution on the host system.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of global NPM packages @openai/codex and @google/gemini-cli. These packages are not official releases from OpenAI or Google and their source is unverified, representing a supply chain risk.
  • [PROMPT_INJECTION]: The skill processes untrusted user data and passes it to secondary models, creating a surface for indirect prompt injection.
  • Ingestion points: User-provided <task description> via the /oh-my-claudecode:ccg command.
  • Boundary markers: None. The content is directly wrapped in double quotes within a shell command.
  • Capability inventory: Execution of shell commands via the omc CLI tool.
  • Sanitization: No sanitization or escaping of shell-sensitive characters is performed on the user input before it is used in a command string.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 20, 2026, 12:42 PM