December 2025 Summaries

Firebase AI Logic has introduced server prompt templates to help developers secure generative AI prompts and update model behavior without requiring client app releases. Templates store system and user instructions, input schemas, and model configurations on Firebase servers, while client applications send only a template ID and validated variable values, reducing prompt exposure and limiting misuse or prompt injection attempts. Built on a partial implementation of the Dotprompt syntax, templates use YAML configuration and Handlebars variables, conditional logic, and iteration, and can be created, tested, edited, and locked through the Firebase console. The feature supports most Gemini and Imagen models, including multimodal inputs, across Firebase AI Logic SDKs for Android, iOS, Flutter, Web, and Unity, with planned additions such as tool support, chat capabilities, and AI monitoring integration.

As of August 2026, Vertex AI is called Agent Platform, and the discussion explains how AI-agent tool designs can expose unauthorized data when they accept user-controlled identifiers, such as a user ID, as parameters. A prompt-injection attack could cause an agent to call a broadly privileged tool with another user’s ID despite application-provided instructions identifying the authenticated user. The recommended approach is to derive identity and other trusted data from authenticated request context, such as verified authorization and app-check tokens, and make tools accept no user-supplied identity parameters. The example uses Genkit context providers to extract a verified UID and lets the notes-retrieval tool access that context directly. It further recommends least-privilege access by having agent workflows call services with the user’s own credentials and restricted endpoints rather than administrative credentials, limiting the potential impact even if tool behavior is manipulated.