examples-run-analysis

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill requires the agent to read and verbatim-quote untruncated log lines from repository logs (and to not ellipsize them), which can force the LLM to output any secrets contained in those logs.