github/openclaw
1
0
Fork 0
mirror of https://github.com/openclaw/openclaw.git synced 2026-04-18 23:27:26 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(slack): add native text streaming support

Browse Source 
Adds support for Slack's Agents & AI Apps text streaming APIs
(chat.startStream, chat.appendStream, chat.stopStream) to deliver
LLM responses as a single updating message instead of separate
messages per block.

Changes:
- New src/slack/streaming.ts with stream lifecycle helpers using
  the SDK's ChatStreamer (client.chatStream())
- New 'streaming' config option on SlackAccountConfig
- Updated dispatch.ts to route block replies through the stream
  when enabled, with graceful fallback to normal delivery
- Docs in docs/channels/slack.md covering setup and requirements

The streaming integration works by intercepting the deliver callback
in the reply dispatcher. When streaming is enabled and a thread
context exists, the first text delivery starts a stream, subsequent
deliveries append to it, and the stream is finalized after dispatch
completes. Media payloads and error cases fall back to normal
message delivery.

Refs:
- https://docs.slack.dev/ai/developing-ai-apps#streaming
- https://docs.slack.dev/reference/methods/chat.startStream
- https://docs.slack.dev/reference/methods/chat.appendStream
- https://docs.slack.dev/reference/methods/chat.stopStream
This commit is contained in:
nathandenherder
2026-02-05 18:14:13 -05:00
parent 82419eaad6
commit 6945fbf100
5 changed files with 340 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
docs/channels/slack.md
View File

@@ -563,6 +563,40 @@ Common failures:
For triage flow: [/channels/troubleshooting](/channels/troubleshooting).
## Text streaming
Slack's [Agents & AI Apps](https://docs.slack.dev/ai/developing-ai-apps) feature includes native text streaming APIs that let your app stream responses word-by-word (similar to ChatGPT) instead of waiting for the full response.
Enable it per-account:
```yaml
channels:
slack:
streaming: true
```
### Requirements
1. **Agents & AI Apps** must be toggled on in your [Slack app settings](https://api.slack.com/apps). This automatically adds the `assistant:write` scope.
2. Streaming only works **within threads** (DM threads, channel threads). Messages without a thread context fall back to normal delivery automatically.
3. Block streaming (`blockStreaming`) is automatically enabled when `streaming` is active so the LLM's incremental output feeds into the stream.
### Behavior
- On the first text block the bot calls `chat.startStream` to create a single updating message.
- Subsequent text blocks are appended via `chat.appendStream`.
- When the reply is complete the stream is finalized with `chat.stopStream`.
- Media attachments (images, files) are delivered as separate messages alongside the stream.
- If a streaming API call fails, the bot gracefully falls back to normal message delivery for the remainder of the response.
### Relevant Slack API methods
| Method | Purpose |
| --------------------------------------------------------------------------------- | ------------------------- |
| [`chat.startStream`](https://docs.slack.dev/reference/methods/chat.startStream) | Start a new text stream |
| [`chat.appendStream`](https://docs.slack.dev/reference/methods/chat.appendStream) | Append text to the stream |
| [`chat.stopStream`](https://docs.slack.dev/reference/methods/chat.stopStream) | Finalize the stream |
## Notes
- Mention gating is controlled via `channels.slack.channels` (set `requireMention` to `true`); `agents.list[].groupChat.mentionPatterns` (or `messages.groupChat.mentionPatterns`) also count as mentions.