competition-ios-runtime

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs capturing and replaying plaintext inputs, headers, tokens, signatures, and Keychain items (preserve/rebuild the smallest replayable request), which requires the agent to handle and potentially emit secret values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill explicitly instructs bypassing pinning/environment checks, hooking Keychain reads, and capturing plaintext tokens/requests — behaviors that directly enable credential theft and covert data exfiltration if misused.