riff

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill contains multiple intentional data-exfiltration / token-use patterns (CLI calls that "use subscription tokens", sending auth tokens to configurable endpoints via query strings/Authorization headers, and broadcast postMessage('*') of error payloads) that create clear avenues for credential misuse or leaking sensitive runtime data to untrusted servers.