ledger
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection due to its processing of untrusted external data combined with its ability to generate significant downstream configurations.
- Ingestion points: The skill ingests infrastructure-as-code (IaC) resource definitions (Terraform, CloudFormation, Pulumi) and cloud utilization metrics (CPU, Memory, GPU) from external environments.
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when parsing these external data sources.
- Capability inventory: The skill generates actionable outputs including IaC modifications for the Scaffold agent, cost anomaly alert rules for the Beacon agent, and CI/CD gate specifications for the Gear agent.
- Sanitization: There are no documented procedures for sanitizing or validating the content of ingested IaC code or metrics to prevent malicious instructions from influencing the agent's logic.
Audit Metadata