gcp-cloud-architect
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from Terraform and YAML configuration files, which is a standard surface for indirect prompt injection.
- Ingestion points: The scripts
scripts/gcp_architecture_validator.py,scripts/gcp_caf_scorer.py, andscripts/gcp_cost_estimator.pyread user-provided file content from disk. - Boundary markers: Absent; findings derived from user files are presented to the agent without explicit delimiters or instructions to ignore embedded directives.
- Capability inventory: The tools are restricted to static analysis (file reading and regex-based pattern matching) and do not possess network exfiltration or system modification capabilities.
- Sanitization: Snippets from the analyzed configuration files are echoed into validation reports without specific filtering for instructional content.
- [SAFE]: No malicious obfuscation, command execution, or unauthorized data exfiltration patterns were detected. The skill is authored by a known vendor and follows security best practices by using isolated local scripts.
Audit Metadata