firebase-ai
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the firebase_ai package via the standard flutter pub add command.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data (text and images), creating a potential attack surface for indirect prompt injection.
- Ingestion points: User input is ingested through generateContent and sendMessage methods as defined in SKILL.md.
- Boundary markers: The skill documentation includes advice on using safety settings and sanitizing inputs to act as boundaries.
- Capability inventory: The skill is designed to facilitate network communication with AI models and perform file reads for multimodal processing.
- Sanitization: The skill provides explicit guidance on implementing content filtering, safety thresholds, and user input sanitization.
Audit Metadata