hook-creator

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

• Third-party content exposure detected (high risk: 0.90). The skill's hooks.json documentation explicitly supports an "http" hook type that can "Call a local or remote HTTP endpoint" and the doc warns that hook output (stdout) is "injected verbatim into the model's context" and PreToolUse/PostToolUse hooks can block or alter tool calls, so remote/untrusted HTTP content could be fetched and directly influence agent behavior.