helicone

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt embeds literal API keys and bearer tokens (e.g., "sk-...", "sk-helicone-xxxx") directly in code and curl examples and instructs them to be included in headers/requests, which requires the agent to handle/output secrets verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The code explicitly routes OpenAI/Anthropic traffic through a third‑party Helicone proxy and attaches Helicone authentication headers plus user/session/ticket identifiers (and logs responses asynchronously), which deliberately sends API keys, request/response payloads, and user-identifying metadata to an external service — representing an intentional data-exfiltration/credential-exposure pattern (no obfuscation or remote‑exec backdoor detected).