oci-order-detail-query

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly tells the agent to ask the user for HOST/PORT/USER/PASSWORD and to create a ~/.duckdbrc containing a CREATE SECRET block that embeds those credentials verbatim, forcing the LLM to handle and output secret values.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required workflow reads outsider-authored free text from the StarRocks/DuckDB query results (e.g., customer/order fields like order_customer_name, sales_name, etc.) into CSV via COPY ... TO 'filename.csv', then inspects/feeds that CSV content into the agent’s LLM context for analysis/presentation.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill includes a runtime installation command that fetches and pipes remote code to a shell using the URL https://install.duckdb.org (curl https://install.duckdb.org | sh), which executes remote code and is a required prerequisite for the skill.