dispatching-parallel-agents

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly says agent prompts may include "credentials" as part of the self-contained context and requires returning evidence (request/response) verbatim, which creates a clear path for secrets (API keys, passwords, tokens) to be included in the LLM output and thus exfiltrated.