git-guardrails-claude-code

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The skill installs a local PreToolUse Bash hook that reads the LLM’s tool_input.command at runtime (user-/outsider-authored text can influence that command), but it does not ingest outsider-authored free text such as chat/email/issues feed content directly—only the generated command string the agent attempts to run is inspected.