gcp-cloud
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as GCP API responses and infrastructure-as-code files, which could potentially contain malicious instructions.
- Ingestion points: The skill ingests data from gcloud, gsutil, and bq CLI outputs, as well as contents of IaC files and Cloud Storage objects.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the ingested data are defined in the instructions.
- Capability inventory: The skill is granted access to high-privilege tools including Bash, Read, Write, Edit, and WebFetch, which allows for shell execution, file modification, and network requests.
- Sanitization: There is no documented validation, escaping, or filtering process for data retrieved from external cloud services before it is used to influence further agent actions.
Audit Metadata