modelslab-chat-generation

Fail

Audited by Snyk on Feb 21, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 0.90). The prompt's examples explicitly require inserting an API key value directly into request payloads and example commands (Python/cURL/OpenAI SDK), so an agent would need to handle and could easily output secret API keys verbatim.
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 21, 2026, 10:07 PM