shift-right-testing
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze production telemetry, incident reports (e.g., INC-2024-001), and logs to generate regression tests and resilience experiments. This ingestion of external, potentially untrusted data creates a surface for indirect instructions if the source data contains adversarial strings.
- Ingestion points:
SKILL.md(Incident Replay task, Synthetic Monitoring, and RUM analysis blocks). - Boundary markers: The current templates do not specify delimiters or instructions to ignore embedded commands in the processed incident data.
- Capability inventory: The skill uses
FleetManager.coordinateandTaskexecution to trigger actions across multiple agents (qe-chaos-engineer,qe-production-intelligence). - Sanitization: The skill does not explicitly define sanitization or filtering logic for the production data before it is processed by the agents.
Audit Metadata