ktx-context-layer-data-agents

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text can be ingested into the LLM context via configured wiki content and external tool artifacts (e.g., context_sources like dbt/Looker and wiki/*), where the runtime ingestion path reads those documents’ body text and uses it for context building.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill documentation explicitly instructs using sudo (e.g., "sudo chown -R $USER:$USER .") to change filesystem ownership and includes commands that start local servers and manage project state, which encourages privilege escalation and modification of system state.