A framework to build Slack apps using Python
Repo slackapi/bolt-python, read from GitHub 21 h ago. slackapi has not joined Take My Repo.
Full Changelog: v1.29.0...v1.30.0
Bring magic to a conversation with say_stream for streaming messages and show loading status with set_status. Now available for app.event and app.message listeners:
@app.event("app_mention") def handle_mention(say_stream, set_status): set_status( status="Thinking...", loading_messages=["Waking up...", "Loading a witty response..."], ) stream = say_stream(buffer_size=100) stream.append(markdown_text="Thinking... :thinking_face:\n\n") stream.append(markdown_text="Here is my response!") stream.stop()
complete & fail utility by @WilliamBergamin in #1390Full Changelog: v1.26.0...v1.27.0
Milestone: https://github.com/slackapi/bolt-python/milestone/95?closed=1
Try the AI Agent Sample app to explore the AI-enabled features and existing Assistant helper:
# Create a new AI Agent app $ slack create slack-ai-agent-app --template slack-samples/bolt-python-assistant-template $ cd slack-ai-agent-app/ # Initialize Python Virtual Environment $ python3 -m venv .venv $ source .venv/bin/activate $ pip install -r requirements.txt # Add your OPENAI_API_KEY $ export OPENAI_API_KEY=sk-proj-ahM... # Run the local dev server $ slack run
After the app starts, send a message to the "slack-ai-agent-app" bot for a unique response.
Loading states allows you to not only set the status (e.g. "My app is typing...") but also sprinkle some personality by cycling through a collection of loading messages.
Bolt Assistant Class usage:
@assistant.user_message def respond_in_assistant_thread( client: WebClient, context: BoltContext, get_thread_context: GetThreadContext, logger: Logger, payload: dict, say: Say, set_status: SetStatus, ): set_status( status="thinking...", loading_messages=[ "Teaching the hamsters to type faster…", "Untangling the internet cables…", "Consulting the office goldfish…", "Polishing up the response just for you…", "Convincing the AI to stop overthinking…", ], )
Web Client SDK usage:
@app.message() def handle_message(client, context, event, message): client.assistant_threads_setStatus( channel_id=channel_id, thread_ts=thread_ts, status="thinking...", loading_messages=[ "Teaching the hamsters to type faster…", "Untangling the internet cables…", "Consulting the office goldfish…", "Polishing up the response just for you…", "Convincing the AI to stop overthinking…", ], ) # Start a new message stream
The chat_stream() helper utility can be used to streamline calling the 3 text streaming methods:
# Start a new message stream streamer = client.chat_stream( channel=channel_id, recipient_team_id=team_id, recipient_user_id=user_id, thread_ts=thread_ts, ) # Loop over OpenAI response stream # https://platform.openai.com/docs/api-reference/responses/create for event in returned_message: if event.type == "response.output_text.delta": streamer.append(markdown_text=f"{event.delta}") else: continue feedback_block = create_feedback_block() streamer.stop(blocks=feedback_block)
Alternative to the Text Streaming Helper is to call the individual methods.
client.chat_startStreamFirst, start a chat text stream to stream a response to any message:
@app.message() def handle_message(message, client, context, event, message): # Start a new message stream stream_response = client.chat_startStream( channel=channel_id, recipient_team_id=team_id, recipient_user_id=user_id, thread_ts=thread_ts, ) stream_ts = stream_response["ts"]
client.chat_appendStreamAfter starting a chat text stream, you can then append text to it in chunks (often from your favourite LLM SDK) to convey a streaming effect:
for event in returned_message: if event.type == "response.output_text.delta": client.chat_appendStream( channel=channel_id, ts=stream_ts, markdown_text=f"{event.delta}" ) else: continue
client.chat_stopStreamLastly, you can stop the chat text stream to finalize your message:
client.chat_stopStream( channel=channel_id, ts=stream_ts, blocks=feedback_block )
Add feedback buttons to the bottom of a message, after stopping a text stream, to gather user feedback:
def create_feedback_block() -> List[Block]: blocks: List[Block] = [ ContextActionsBlock( elements=[ FeedbackButtonsElement( action_id="feedback", positive_button=FeedbackButtonObject( text="Good Response", accessibility_label="Submit positive feedback on this response", value="good-feedback", ), negative_button=FeedbackButtonObject( text="Bad Response", accessibility_label="Submit negative feedback on this response", value="bad-feedback", ), ) ] ) ] return blocks @app.message() def handle_message(message, client): # ... previous streaming code ... # Stop the stream and add feedback buttons feedback_block = create_feedback_block() client.chat_stopStream( channel=channel_id, ts=stream_ts, blocks=feedback_block )
chat_postMessage supports markdown_textresponse = client.chat_postMessage( channel="C111", markdown_text=markdown_content, )
Learn more in slackapi/python-slack-sdk#1718
📚 https://docs.slack.dev/reference/block-kit/blocks/markdown-block/
from slack_sdk.models.blocks import MarkdownBlock ... @app.message("hello") def message_hello(say): say( blocks=[ MarkdownBlock(text="**lets's go!**"), ], text="let's go!", )
Learn more in slackapi/python-slack-sdk#1748
Add support for the workflows.featured.{add|list|remove|set} methods:
app.client.workflows_featured_add(channel_id="C0123456789", trigger_ids=["Ft0123456789"]) app.client.workflows_featured_list(channel_ids="C0123456789") app.client.workflows_featured_remove(channel_id="C0123456789", trigger_ids=["Ft0123456789"]) app.client.workflows_featured_set(channel_id="C0123456789", trigger_ids=["Ft0123456789"])
Learn more in slackapi/python-slack-sdk#1712