deep-security-scan

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt explicitly instructs the agent to accept and forward identifiers/tokens (e.g., sessionId, scanId, handoffClaimToken) verbatim between tool calls, which requires handling secret-like values and creates an exfiltration risk if those values are emitted in output.