chore: remove slack from openai sample (#83)

The sample uses a URL parameter for input, which simplifies the
implementation but might not be realistic for more complex prompts. I
have two options for addressing this:

- **Option 1**: Document the current approach, informing users of its
limitations with URL parameters for input.
- **Option 2**: Rewrite the sample to accept input through the request
body, which would introduce slightly more complexity but offer greater
flexibility.

Additionally, I’ve adjusted how user input influences the prompt.
Previously, with the URL parameter approach, the input wasn’t impacting
ChatGPT’s responses as expected. Now, users can directly enter their own
prompt and see its effect on the output more clearly. If we switch to
using a JSON body, we might be able to revert to the previous method of
prompting ChatGPT, as longer inputs could have a more noticeable impact.

---------

Co-authored-by: Daniel Abraham <[email protected]>

samples/openai_chatgpt/README.md

@@ -3,26 +3,36 @@
 This [AutoKitteh](https://github.com/autokitteh/autokitteh) project
 demonstrates integration with [ChatGPT](https://chat.openai.com).
 
-The file [`program.star`](./program.star) implements a single entry-point
+The file [`program.py`](./program.py) implements a single entry-point
 function, which is configured in the [`autokitteh.yaml`](./autokitteh.yaml)
-manifest file as the receiver of Slack `slash_command` events.
+manifest file as the receiver of HTTP events.
 
 It sends a couple of requests to the ChatGPT API, and sends the responses
 back to the user over Slack, as well as ChatGPT token usage stats.
 
 API details:
 
 - [OpenAI developer platform](https://platform.openai.com/)
-- [Go client API](https://pkg.go.dev/github.com/sashabaranov/go-openai)
+- [Python client API](https://github.com/openai/openai-python)
 
-This project isn't meant to cover all available functions and events. it
+This project isn't meant to cover all available functions and events. It
 merely showcases a few illustrative, annotated, reusable examples.
 
 ## Instructions
 
-1. Follow instructions [here](https://platform.openai.com/docs/quickstart) to setup your OpenAI API account.
+1. Follow instructions [here](https://platform.openai.com/docs/quickstart) to set up your OpenAI API account.
 
-2. Via the `ak` CLI tool, or the AutoKitteh WebUI, initialize the OpenAI connection and provide API key generated in step 1. 
+2. Via the `ak` CLI tool, or the AutoKitteh WebUI, initialize the OpenAI connection and provide the API key generated in step 1.
 
-3. Via the `ak` CLI tool, or the AutoKitteh VS Code extension, deploy the
-   `autokitteh.yaml` manifest file
+3. Via the `ak` CLI tool, or the AutoKitteh VS Code extension, deploy the `autokitteh.yaml` manifest file.
+
+4. Once deployed, the program is ready to receive HTTP POST requests. You can test the program by sending a POST request to the endpoint as shown below:
+
+   ```shell
+   curl -X POST "http://localhost:9980/webhooks/<webhook_slug>" -H "Content-Type: text/plain" -d "Meow"
+   ```
+
+5. You can modify the request body in the curl command to send custom text and observe how ChatGPT responds with dynamic content.
+
+> [!NOTE]
+> The [`autokitteh.yaml`](autokitteh.yaml) manifest file is set up to filter incoming HTTP requests. You can modify or remove this filter as needed.

samples/openai_chatgpt/autokitteh.yaml

@@ -9,10 +9,9 @@ project:
   connections:
     - name: chatgpt_conn
       integration: chatgpt
-    - name: slack_conn
-      integration: slack
   triggers:
-    - name: slack_slash_command
-      connection: slack_conn
-      event_type: slash_command
-      call: program.py:on_slack_slash_command
+    - name: on_http_get
+      type: webhook
+      event_type: post
+      filter: data.headers["Content-Type"] == "text/plain"
+      call: program.py:on_http_get

samples/openai_chatgpt/program.py

@@ -1,39 +1,40 @@
 """This program demonstrates AutoKitteh's OpenAI ChatGPT integration.
 
-This program implements a single entry-point function, which is
-configured in the "autokitteh.yaml" manifest file as the receiver
-of Slack "slash_command" events.
+The program implements a single entry-point function, which is
+configured in the "autokitteh.yaml" manifest file to receive HTTP GET requests.
 
-It sends a couple of requests to the ChatGPT API, and sends the responses
-back to the user over Slack, as well as ChatGPT token usage stats.
+It sends a couple of requests to the ChatGPT API, and prints the responses
+in the AutoKitteh session log, along with ChatGPT token usage stats.
 
 API details:
 - OpenAI developer platform: https://platform.openai.com/
-- Go client API: https://pkg.go.dev/github.com/sashabaranov/go-openai
+- Python client library: https://github.com/openai/openai-python
 
 This program isn't meant to cover all available functions and events.
 It merely showcases various illustrative, annotated, reusable examples.
-
 """
 
-from autokitteh import openai, slack
+from autokitteh.openai import openai_client
 
 MODEL = "gpt-4o-mini"
 
-chatgpt_client = openai.openai_client("chatgpt_conn")
-slack_client = slack.slack_client("slack_conn")
+chatgpt_client = openai_client("chatgpt_conn")
 
 
-def on_slack_slash_command(event):
-    """https://api.slack.com/interactivity/slash-commands
+def on_http_get(event):
+    """
+    Entry-point function for handling HTTP GET requests in this workflow.
 
-    To use the slash command, simply type `/command-name` in the Slack message input,
-    where `command-name` is the name you have assigned to the command in your app.
-    This command does not require any additional text or arguments.
+    Example usage:
+    - URL: "http://localhost:9980/webhooks/<webhook_slug>"
+    - Curl command:
+      curl -X POST "<URL>" -H "Content-Type: text/plain" -d "Meow"
 
     Args:
-        event: Slack event data.
+        event: The HTTP event containing request data.
     """
+    body = event.data.body.bytes.decode("utf-8")
+    print(body)
 
     # Example 1: trivial interaction with ChatGPT.
     msg = {"role": "user", "content": "Meow!"}
@@ -51,13 +52,11 @@ def on_slack_slash_command(event):
     ]
     msgs = [
         {"role": "system", "content": contents[0]},
-        {"role": "user", "content": contents[1]},
-        {"role": "user", "content": event.data["text"]},
+        {"role": "user", "content": body or contents[1]},
     ]
 
     resp = chatgpt_client.chat.completions.create(model=MODEL, messages=msgs)
 
-    id = event.data["user_id"]
     for choice in resp.choices:
-        slack_client.chat_postMessage(channel=id, text=choice.message.content)
-    slack_client.chat_postMessage(channel=id, text=f"Usage: `{str(resp.usage)}`")
+        print(choice.message.content)
+    print(f"Usage: `{resp.usage}`")