firebase-ai-logic
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructs the user to install official packages from well-known and trusted organizations, specifically
firebase-toolsandfirebasefrom Google, and@anthropic-ai/sdkfrom Anthropic. - [SAFE]: Security best practices are emphasized, including the use of environment variables (
.env.local) for secret management and the use of placeholders (YOUR_API_KEY) to prevent accidental hardcoding of credentials. - [SAFE]: The skill explicitly defines constraints requiring developers to perform user input validation and avoid sending sensitive data to AI models, which mitigates risks associated with indirect prompt injection.
Audit Metadata