Update changelog for version 0.1.16, add configurable wake-word support, optional always-on voice mode, and websocket settings endpoint; enhance chat card voice recognition and context handling.
This commit is contained in:
@@ -44,6 +44,18 @@ Configure this in Home Assistant:
|
||||
2. Open the **Expose** tab
|
||||
3. Select which entities are exposed to **Assist**
|
||||
|
||||
## Advanced options
|
||||
|
||||
Open **Settings → Devices & Services → OpenClaw → Configure** to tune:
|
||||
|
||||
- Include exposed entities context
|
||||
- Max context characters
|
||||
- Context overflow strategy (`truncate` or `clear`)
|
||||
- Enable tool calls (execute HA services from model tool-call responses)
|
||||
- Require wake word
|
||||
- Wake word text (for example: `hey openclaw`)
|
||||
- Always-on voice mode (continuous listening while the card is open)
|
||||
|
||||
## Chat Card
|
||||
|
||||
The chat card is registered automatically when the integration loads — no
|
||||
|
||||
Reference in New Issue
Block a user