langfuse-integration

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). No obfuscated backdoors or remote-exec constructs were found, but the content explicitly and repeatedly instructs automated instrumentation that will send sensitive pharmaceutical data (user IDs/emails, session IDs, URS contents, compliance metadata) and uses environment-held API keys to a specific external Langfuse Cloud project — a high-risk data-exfiltration / supply-chain exposure if the remote project or keys are untrusted.