> [!WARNING] > **Preview**: Using server prompt templates is a feature that's in Preview, which means that it isn't subject to any SLA or deprecation policy and could change in backwards-incompatible ways.
This page describes some [best practices](https://firebase.google.com/docs/ai-logic/server-prompt-templates/best-practices-and-considerations.md.txt#best-practices) and considerations for server prompt templates, including [not-yet-supported features](https://firebase.google.com/docs/ai-logic/server-prompt-templates/best-practices-and-considerations.md.txt#not-yet-supported). Note that many of these features will become available as server prompt templates evolve, so check the [release notes](https://firebase.google.com/support/releases.md.txt) for updates! ## Best practices Many of these best practices are also detailed in [Manage your templates](https://firebase.google.com/docs/ai-logic/server-prompt-templates/manage-templates.md.txt). ### Version your templates - Create template IDs appended with a version that uses [semantic versioning (semver)](https://semver.org/). - Use [Firebase Remote Config](https://firebase.google.com/docs/ai-logic/server-prompt-templates/versioning-with-remote-config.md.txt) so that you can easily change the template and other values in your request. ### Protect your template - Avoid editing templates that are used in production unless you've thoroughly tested the change. - Lock your template before going to production. - Locking a template acts as protection against unintentional editing, but locking *does **not** entirely block editing*. A project member with the appropriate permissions can always unlock a template to edit it. - We recommend locking templates that are actively being used by code -- especially production code. > [!NOTE] > **Note:** If you've enabled [AI monitoring](https://firebase.google.com/docs/ai-logic/monitoring.md.txt#ai-monitoring-in-console), then you can see which requests use server prompt templates. You can find this information when you click into the **Details** of a trace in the [**Trace Explorer**](https://console.cloud.google.com/traces/explorer) in the Google Cloud console (and it's coming soon to the Firebase console!). ### Protect usage of the Gemini API - Write strong [input validation](https://firebase.google.com/docs/ai-logic/server-prompt-templates/syntax-and-examples.md.txt#gemini-input-validation-and-schema) for your input variables, which can help with the following: - Can help protect against prompt injection. - Can help ensure requests succeed and responses are as expected. - Enforce [template-only mode](https://firebase.google.com/docs/ai-logic/server-prompt-templates/template-only-mode.md.txt) for your Firebase project so that all Gemini API requests via Firebase AI Logic must use a server prompt template. ## Features not yet supported by server prompt templates Many of these features will become available as server prompt templates evolve, so check the [release notes](https://firebase.google.com/support/releases.md.txt) for updates! ### Capabilities of Firebase AI Logic not yet supported by templates The following standard capabilities from Firebase AI Logic are not yet supported through server prompt templates: - Bidirectional streaming (Gemini Live API) - Hybrid or on-device inference - Constraining the output to a list of enums Also note that if you use AI monitoring in the Firebase console, the template ID is not yet populated in any record. ### Common elements of Dotprompt not yet supported by templates Server prompt templates ***will likely soon*** support the following common elements of Dotprompt: - Using the [JSON schema specification](https://json-schema.org/) for input and output schemas in your frontmatter. - Note that this spec isn't supported in non-server template requests either. We currently only support the [OpenAPI schema specification](https://swagger.io/specification/). - Declaring a wildcard field definition using `*`. - Using `@key` or generally iterating over the fields of an object input (which is the case where `@key` is relevant). - Using `@root`, which lets you reference the root variable context regardless of current `this`. Server prompt templates ***will likely not*** support the following common elements of Dotprompt: - Using partials, which are reusable template snippets that can be included in other templates. ## Other considerations - By default, when you use the guided UI in the Firebase console, we provision the template in all [supported regions for Firebase AI Logic](https://firebase.google.com/docs/ai-logic/locations.md.txt?api=vertex#available-locations). If you use the Agent Platform Gemini API (formerly Vertex AI) and if your use case requires location-based restrictions, you can [specify the location for your template](https://firebase.google.com/docs/ai-logic/server-prompt-templates/advanced-workflows.md.txt) using the REST API. - If you want to [provide a server prompt template as a file](https://firebase.google.com/docs/ai-logic/server-prompt-templates/advanced-workflows.md.txt) (instead of using the Firebase console's guided UI), you can use the REST API. Note that server prompt templates do *not* support schema defined in app code and passed into the template. - Even though the template is on the server, it cannot ***directly*** interact with your Firebase project's other server-side resources (like a database) *except for Cloud Storage for Firebase URLs* (which can be provided as input variables).