claude-code-design-guide

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The guide includes patterns that pass tool parameters and tool results (via JSON.stringify) into LLM prompts and ask the LLM to "fix" or return parameters (e.g., executeToolWithRetry and streaming tool-call handling), which would force the model to receive and potentially emit secret values verbatim if those params/results contain credentials.