firebase-ai-logic
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted user data into the agent context and lacks explicit boundary markers or sanitization logic.
- Ingestion points: The
generateText,analyzeImage, andsendMessagefunctions inreferences/usage_patterns_web.mdaccept unvalidated user prompts and file data. - Boundary markers: None are present in the provided code snippets to delimit user content from system instructions.
- Capability inventory: The skill facilitates network operations via the Firebase SDK to communicate with Gemini models.
- Sanitization: There is no evidence of input validation, escaping, or filtering of external content before it is interpolated into model calls.
Audit Metadata