kubesense-alerts
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from existing KubeSense alerts and external Datadog monitor definitions to generate or modify alert rule configurations.
- Ingestion points: Data is ingested from the environment via tool outputs from
get-alert-detailsandlist-alert-rules, and from user-provided input when migrating Datadog monitors or importing JSON rules. - Boundary markers: No specific delimiters or safety instructions are defined to separate ingested rule content from agent instructions.
- Capability inventory: The skill allows for the creation of alert rules through the
create-alerttool and provides guidance on manual bulk JSON imports. - Sanitization: While the
validate-alert-jsontool provides schema and structural verification, it does not sanitize natural language fields such as alert names or descriptions for malicious instructions.
Audit Metadata