google-agents-cli-scaffold
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user requirements and project specifications to configure and execute
agents-clicommands. While this introduces a surface for indirect prompt injection, it is managed through explicit validation rules and an interactive workflow. - Ingestion points: User-provided project names, architecture choices (RAG, A2A), and deployment requirements gathered during Phase 0 in
SKILL.md. - Boundary markers: None explicitly defined in the prompt templates, but the instructions require the agent to clarify requirements before execution.
- Capability inventory: The skill uses the
agents-clibinary to perform file system operations (scaffolding), infrastructure provisioning (infra datastore), and data ingestion. - Sanitization: The skill provides explicit constraints for project names (length and character limitations) to prevent basic command injection via parameter passing.
- [EXTERNAL_DOWNLOADS]: The skill directs the installation of the
google-agents-clitool viauv. This tool is provided by a well-known organization and is a standard dependency for the described development workflow.
Audit Metadata