grill-with-docs

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill requires scanning the repo and inserting short code snippets from files into questions without any instruction to redact secrets, so if secrets (API keys, tokens, passwords) exist in the codebase the agent may be forced to reproduce them verbatim—creating an exfiltration risk.