langfuse

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs obtaining API keys and includes example CLI commands that embed secrets as environment variables (e.g., LANGFUSE_SECRET_KEY=sk-...), which encourages the LLM to ask for and/or emit secret values verbatim in generated commands—an insecure pattern.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill configures LANGFUSE_HOST (https://cloud.langfuse.com) as the MCP runtime endpoint and includes tools like get_prompt/get_prompt_unresolved that fetch prompt content at runtime, which can directly control agent instructions fetched from that URL.