multi-model-orchestrator

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs users/agents to "copy" and "cat" the handoff YAML into agent prompts (and to have Codex read the file), which will cause any secrets present in the handoff to be included verbatim in LLM inputs/outputs, creating an exfiltration risk.