delegation-principle

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt mandates reading state files and "pass complete context to subagent" via the Task tool, which effectively requires the agent to include whatever secret values exist in that context verbatim in its outgoing tool payloads (exposing secrets), even though it doesn't explicitly tell the agent to ask for keys.