insideout

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill forces the agent to relay tool/Riley responses verbatim and exposes tools (e.g., tfoutputs, credawait) that can return credentials or sensitive outputs, so the LLM may be required to output secret values verbatim despite guidance to avoid including them.