speak-ci-integration

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No attempts to override agent behavior or bypass safety guidelines were detected.
  • [DATA_EXFILTRATION]: API credentials are handled via environment variables and GitHub Secrets. Network communication is directed solely to the service's official domain (api.speak.com).
  • [COMMAND_EXECUTION]: Commands are limited to standard CI/CD operations such as API health checks and running test suites.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes verified GitHub Actions and reputable libraries like Vitest and Zod from the NPM registry.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 10:28 PM