agent-platform-prompt-management

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the official google-cloud-aiplatform and google-genai libraries from the standard Python package registry.- [INDIRECT_PROMPT_INJECTION]: The skill manages prompt templates and associated data variables, establishing a surface for indirect prompt injection if malicious content is stored or processed.
  • Ingestion points: Template strings and user-provided variables processed in SKILL.md and references/create.md.
  • Boundary markers: The instructions utilize standard curly brace delimiters (e.g., {text} or {{variable}}) to define template placeholders.
  • Capability inventory: The skill provides access to SDK methods for listing, retrieving, creating, and deleting managed prompt resources.
  • Sanitization: The skill relies on the underlying Vertex AI SDK's template interpolation without implementing additional explicit sanitization or filtering logic.- [DYNAMIC_EXECUTION]: The skill generates Python scripts from predefined templates to perform prompt lifecycle operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 05:41 AM
Security Audit — agent-trust-hub — agent-platform-prompt-management