antigravity-cli

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill documentation intentionally promotes bypassing permission checks (--dangerously-skip-permissions), injecting local file contents and directories into prompts/agent context, and installing/running external plugins and MCP servers — patterns that enable data exfiltration, remote code execution, and supply‑chain abuse.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill repeatedly instructs automation to use the explicit --dangerously-skip-permissions flag (and shows examples), which encourages bypassing the CLI's permission prompts and thus weakens a security barrier for automated agents even though it doesn't directly request sudo or system-wide changes.