multi-ai-collab

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to read file contents and inline them into prompts/CLI commands (e.g., using $(cat path/to/file) or the Read tool to replace [CODE_CONTENT]), which would cause any embedded secrets (API keys, tokens, passwords) to be emitted verbatim in the agent's output.