google-gemini-embeddings

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a high-quality guide for developers, providing code examples that interact with official Google and Cloudflare APIs in a transparent and secure manner.\n- [COMMAND_EXECUTION]: Includes a helper script scripts/check-versions.sh that executes standard npm view commands to help users maintain their environment. This is a local development utility and does not represent a runtime security risk.\n- [CREDENTIALS_UNSAFE]: Correctly implements secret management by using environment variables (GEMINI_API_KEY) and Cloudflare Worker secrets. Placeholders are used in templates, ensuring no real credentials are leaked.\n- [EXTERNAL_DOWNLOADS]: Recommends the installation of the official @google/genai package from the npm registry. These references are to well-known and trusted organizations.\n- [PROMPT_INJECTION]: The skill utilizes standard RAG (Retrieval-Augmented Generation) patterns. While it interpolates external data into prompts, it uses clear delimiters and lacks any privileged system capabilities that could be exploited via indirect injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:49 AM