dt-obs-gcp
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests GCP resource metadata, such as resource names, labels, and configuration JSON, through Dynatrace Smartscape. This data acts as a potential surface for indirect prompt injection if an attacker controls resource properties in the monitored GCP environment.\n
- Ingestion points: DQL queries in
SKILL.mdand across all reference files in thereferences/directory that retrieve resource properties.\n - Boundary markers: Absent; query templates interpolate resource data directly into the agent's context without delimiters or instructions to ignore embedded commands.\n
- Capability inventory: The skill uses
smartscapeNodesandtraversecommands for resource discovery and topology mapping within Dynatrace, as documented inSKILL.md.\n - Sanitization: Absent; there is no evidence of filtering or validation of the ingested GCP resource metadata before it is processed by the agent.
Audit Metadata