for-each-line

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill sends each line's raw content verbatim into prompts (and shows the test response for the first line) and writes model outputs to a file, so any secrets in the lines (API keys, tokens, passwords) would be sent to and potentially echoed back by the LLM, enabling exfiltration.