searching-media

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to confirm and apply the complete image URL (including query parameters) returned by tools, which may contain authentication tokens or other secrets that the LLM would need to output verbatim.

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

• Third-party content exposure detected (low risk: 0.10). The runtime workflow first asks the user to choose a trusted search source, then searches internal Salesforce CMS/Data Cloud via tools using either structured keyword/taxonomy extraction or the user’s query as-is; it does not automatically ingest outsider-authored free text streams like inboxes, issues, or monitored feeds.