mirror of
https://github.com/openclaw/openclaw.git
synced 2026-05-06 15:20:44 +00:00
docs(plugins): expand google meet realtime consult docs
This commit is contained in:
@@ -169,7 +169,7 @@ Set config under `plugins.entries.google-meet.config`:
|
||||
realtime: {
|
||||
provider: "openai",
|
||||
model: "gpt-realtime",
|
||||
instructions: "You are joining a private Google Meet as Peter's OpenClaw agent. Keep replies brief unless asked.",
|
||||
instructions: "You are joining a private Google Meet as my OpenClaw agent. Keep replies brief unless asked.",
|
||||
toolPolicy: "safe-read-only",
|
||||
providers: {
|
||||
openai: {
|
||||
@@ -207,6 +207,29 @@ Agents can use the `google_meet` tool:
|
||||
Use `action: "status"` to list active sessions or inspect a session ID. Use
|
||||
`action: "leave"` to mark a session ended.
|
||||
|
||||
## Realtime agent consult
|
||||
|
||||
Chrome realtime mode is optimized for a live voice loop. The realtime voice
|
||||
provider hears the meeting audio and speaks through the configured audio bridge.
|
||||
When the realtime model needs deeper reasoning, current information, or normal
|
||||
OpenClaw tools, it can call `openclaw_agent_consult`.
|
||||
|
||||
The consult tool runs the regular OpenClaw agent behind the scenes with recent
|
||||
meeting transcript context and returns a concise spoken answer to the realtime
|
||||
voice session. The voice model can then speak that answer back into the meeting.
|
||||
|
||||
`realtime.toolPolicy` controls the consult run:
|
||||
|
||||
- `safe-read-only`: expose the consult tool and limit the regular agent to
|
||||
`read`, `web_search`, `web_fetch`, `x_search`, `memory_search`, and
|
||||
`memory_get`.
|
||||
- `owner`: expose the consult tool and let the regular agent use the normal
|
||||
agent tool policy.
|
||||
- `none`: do not expose the consult tool to the realtime voice model.
|
||||
|
||||
The consult session key is scoped per Meet session, so follow-up consult calls
|
||||
can reuse prior consult context during the same meeting.
|
||||
|
||||
## Notes
|
||||
|
||||
Google Meet's official media API is receive-oriented, so speaking into a Meet
|
||||
@@ -214,12 +237,6 @@ call still needs a participant path. This plugin keeps that boundary visible:
|
||||
Chrome handles browser participation and local audio routing; Twilio handles
|
||||
phone dial-in participation.
|
||||
|
||||
Realtime mode gives the voice model one tool, `openclaw_agent_consult`, unless
|
||||
`realtime.toolPolicy` is `none`. The tool asks the normal OpenClaw agent for a
|
||||
concise spoken answer, using recent meeting transcript as context. With
|
||||
`safe-read-only`, the consult run is limited to read/search/memory tools. With
|
||||
`owner`, it inherits the normal agent tool policy.
|
||||
|
||||
Chrome realtime mode needs either:
|
||||
|
||||
- `chrome.audioInputCommand` plus `chrome.audioOutputCommand`: OpenClaw owns the
|
||||
@@ -231,3 +248,7 @@ Chrome realtime mode needs either:
|
||||
For clean duplex audio, route Meet output and Meet microphone through separate
|
||||
virtual devices or a Loopback-style virtual device graph. A single shared
|
||||
BlackHole device can echo other participants back into the call.
|
||||
|
||||
`googlemeet leave` stops the command-pair realtime audio bridge for Chrome
|
||||
sessions. For Twilio sessions delegated through the Voice Call plugin, it also
|
||||
hangs up the underlying voice call.
|
||||
|
||||
Reference in New Issue
Block a user