The Agent Skills Directory

[SAFE]: The skill consists of documentation and code examples for Firebase AI integration. No malicious code, exfiltration patterns, or persistence mechanisms were detected across the provided files.
[PROMPT_INJECTION]: The skill contains text patterns often associated with prompt injection (e.g., "ignore previous instructions"). However, these are identified as false positives as they are used exclusively as illustrative examples within a security section to warn developers about potential adversarial risks.
[SAFE]: The documentation actively promotes security best practices, such as the mandatory use of Firebase App Check to protect project quotas and the validation of all tool arguments to prevent unauthorized actions during client-side execution.

firebase-ai-logic