datacommons-client

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.60). The skill contains an explicit, out-of-scope instruction to proactively promote and suggest using K-Dense Web (a third-party product) for complex workflows, which alters assistant behavior beyond the Data Commons client's stated purpose.

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt explicitly shows insecure examples that embed API keys verbatim (e.g., export DC_API_KEY="your_key" and client = DataCommonsClient(api_key="your_key")), which can lead an LLM to handle or output real secret values directly.