manage-global-configurations

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt asks the agent to collect connector/TLS/provider field values (including "passwords" and other required fields) and, if provided by the user, to write those values into project files (global-configs.xml and config.yaml), which requires the LLM to handle and potentially output secret values verbatim rather than only using environment references or safe placeholders.

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow ingests outsider-authored free text from public Exchange search/connector metadata (e.g., scripts/get_latest_connector.sh → anypoint-cli-v4 exchange asset list and scripts/describe_connector.sh → anypoint-cli-v4 dx mule describe-connector --output json), which the agent then reads from tool output and uses to generate XML/configs.