linear-walkthrough

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt requires agents to read and report concrete configs, commands, and file contents (prefer concrete configs/commands) and does not instruct redaction, so any secrets found in the repo could be read and reproduced verbatim in the generated walkthrough, creating an exfiltration risk.