genkit-production-expert

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill scaffolding script downloads official packages from npm (genkit, @genkit-ai/googleai, @genkit-ai/vertexai) and PyPI (firebase-genkit). These are legitimate libraries belonging to the Google and Firebase AI development ecosystem.
  • [COMMAND_EXECUTION]: The scripts/init-genkit.sh script executes standard commands for project initialization, such as npm init, npm install, and pip install. The script utilizes safe shell practices, including variable quoting and strict error handling with set -euo pipefail.
  • [DATA_EXFILTRATION]: No malicious data access or exfiltration patterns were identified. The skill documentation emphasizes secure practices like environment variable management and API key rotation.
  • [PROMPT_INJECTION]: Instructions and metadata were scanned for injection patterns, such as behavior overrides or safety bypasses, and none were found.
  • [SAFE]: The skill demonstrates a strong security posture by providing documentation on schema validation, input sanitization, and comprehensive error handling for production environments.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 11:48 PM