You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on May 22, 2026. It is now read-only.
Copy file name to clipboardExpand all lines: _includes/docs/mobile/qr-code-settings-before-v1.7.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -51,7 +51,7 @@ Find the app package name **applicationId** in **build.gradle** file located at
51
51
52
52
If your app distributed via Google Play you need to use SHA-256 certificate fingerprint located in your developer account under **Release> Setup > App Integrity> App Signing tab:**
Copy file name to clipboardExpand all lines: _includes/docs/mobile/release.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -12,7 +12,7 @@ Your app package name and app name will come preconfigured from your Mobile Cent
12
12
13
13
If you need to change some settings, we recommend updating them in the Mobile Center and redownloading the configuration file from the Mobile Center. This ensures consistency and prevents configuration conflicts.
**Note:** Always use the Mobile Center to manage your app configuration rather than manually editing configuration files. This approach ensures all settings remain synchronized and reduces the risk of errors.
Copy file name to clipboardExpand all lines: _includes/docs/user-guide/rule-engine-2-0/external-nodes.md
+16-16Lines changed: 16 additions & 16 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -24,9 +24,9 @@ When you select a model from the list, all of its settings defined on that page
24
24
For convenience, you can also add a new model directly from this interface by clicking the "Create new" button, which will open the model configuration form.
-**User prompt** - contains the specific, immediate task or question you want the AI to answer. This is the main input for the model, which will be processed according to the rules set in the system prompt. This field is required, cannot be blank, and also has a maximum length of 10,000 characters.
44
44
> **Example**: "Based on these readings, is the soil moisture optimal for planting corn? Readings: `$[*]`"
You can make both prompts dynamic (incorporate data from an incoming message) using [templatization](/docs/{{docsPrefix}}user-guide/templatization/). This is the recommended approach when you need to provide the context of a message to the AI model.
@@ -78,24 +78,24 @@ Example: If the prompt is `Device context: ${*}` and the incoming metadata is:
78
78
-**Text** - most flexible format, supported by all models. In this mode, the AI can generate free-form text without any structural constraints. While the output is not guaranteed to be in any specific format, you can still guide the model to produce structured data (like JSON) through clear instructions in your prompts.
-**JSON** - instructs the model to generate a response that is always a syntactically valid JSON. The model decides on the JSON structure itself based on the context of the prompt.
-**JSON Schema** - forces the model to generate a JSON that strictly conforms to a specific structure you define using a [JSON Schema](https://json-schema.org/). This is useful for ensuring reliable output.
@@ -110,17 +110,17 @@ Example: If the prompt is `Device context: ${*}` and the incoming metadata is:
110
110
111
111
-**Timeout** - the maximum time the node will wait for a response from the AI model before the request fails. The value must be from 1 second to 10 minutes (600 seconds).
> **Note**: Be careful when setting this value. Complex tasks or slower models may require more time to generate a response. A timeout that is too low for your use case might cause requests to fail unnecessarily.
- If enabled, the incoming message is acknowledged immediately. A new message is created to carry the AI's response and is then added to the queue for processing by the next node. This is useful for long-running AI requests to prevent message processing timeouts.
126
126
- If disabled, the original incoming message is transformed. Its payload is replaced with the AI's response, and this message is passed to the next node.
0 commit comments