cloudrun-development
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve processing third-party application code, Dockerfiles, and configuration files, which represents an attack surface for indirect prompt injection.
- Ingestion points: The skill processes existing or third-party applications and container images (e.g., from Docker Hub or GitHub repositories) as part of the deployment workflow in
SKILL.mdandreferences/vpc-and-database.md. - Boundary markers: It references a mandatory 'Sensitive Runtime Data Protection' protocol and explicitly warns against echoing system-injected headers (like
x-cloudbase-context) or environment variables in responses. - Capability inventory: The agent has access to deployment tools (
manageCloudRun), log inspection (queryCloudRun), and shell commands for container image management. - Sanitization: The skill emphasizes using environment-based API keys issued through platform-controlled APIs rather than local login state files, and it requires following VPC security group rules to isolate network traffic.
Audit Metadata