google-cicd-release-orchestration
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the
gcloudCLI, specificallygcloud deploy apply,gcloud deploy releases create, andgcloud deploy promote. These commands are essential for its purpose of CICD release orchestration and are performed based on user-provided project and region configuration. - [INDIRECT_PROMPT_INJECTION]: The skill acts on external data provided by the user, such as existing runtime manifests for Cloud Run or GKE, which creates a potential surface for instructions embedded in that data to influence agent behavior.
- Ingestion points: User-supplied runtime manifests and deployment parameters (SKILL.md).
- Boundary markers: The skill does not define explicit delimiters or warnings for the agent to ignore instructions that might be contained within the user-provided manifests or project metadata.
- Capability inventory: The skill possesses the ability to execute shell commands (
gcloud) and write configuration files to the local file system (SKILL.md). - Sanitization: No specific sanitization or validation logic is defined for the content of user-provided manifests before they are used to generate Skaffold or Cloud Deploy configurations.
Audit Metadata