google-cloud-agent-sdk-master

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions in references/SKILL.full.md to install the agent-starter-pack and google-cloud-aiplatform packages using pip and uvx. These tools are associated with official Google Cloud repositories and well-known registries.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of Retrieval-Augmented Generation (RAG) agents and uses tools like WebSearch and WebFetch. This creates a surface where external, untrusted content from web searches or document stores could influence the agent's behavior.
  • Ingestion points: Untrusted data enters via WebSearch, WebFetch, and RAG tools (VectorSearchTool, VertexAISearchTool) as described in SKILL.md and references/SKILL.full.md.
  • Boundary markers: The instructions do not explicitly specify boundary markers for external content, though they recommend using specific models like Gemini.
  • Capability inventory: The skill uses Read, WebFetch, WebSearch, and Grep tools.
  • Sanitization: No specific sanitization or filtering logic is provided for the content retrieved from external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 09:56 AM