mundi-orch-multi-llm-route

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The I/O contract explicitly says the skill reads an api_keys.json containing "all provider keys" and the dispatch steps reference using provider SDKs or curl patterns, which forces the agent to access secrets and (in practice) embed them into outgoing requests/commands, creating an exfiltration risk.