context-blocks

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The API permits un-sandboxed dynamic Python expressions (Context(expr=...)) and exposes internal runtime/state blocks (e.g. self, state) that can be set from multiple entry points (strategies, user-supplied contexts), creating high-risk vectors for remote code execution and sensitive data exfiltration if untrusted parties can supply context values.