refinery

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The refinery is instructed to read and rewrite source files (and "show your work" and write revised code back), so if those artifacts contain embedded secrets or keys the agent would read and emit them verbatim — the prompt gives no safe-handling constraints like avoiding printing secrets or using environment variables.