Adding Podcast summarizer Agent

siddartha-10 commented 1 week ago

User description

Summarizing a whole podcast form youtube and send it as a slack message to a channel using composio and crewai. Simply input the YouTube podcast URL and your preferred Slack channel—The Crew handles the rest, summarizing the content and delivering it seamlessly.

PR Type

Enhancement, Documentation

Description

Implemented PodSumCrew class to handle podcast summarization and Slack messaging.
Added tools for audio transcription and Slack integration.
Created a Streamlit app for user input and triggering the summarization process.
Configured agents and tasks for the summarization and messaging processes.
Added MIT License and README with project description and setup instructions.
Included project dependencies in requirements.txt.

Changes walkthrough 📝

Relevant files

Enhancement

4 files

Podcast_Summarizer_AI_Agent.py `Implement podcast summarization and Slack messaging agents` cookbook/Podcast_summarizer_Agents/Podcast_Summarizer_AI_Agent.py Added `PodSumCrew` class to handle podcast summarization and Slack messaging. Configured agents and tasks for summarizing podcasts and sending messages. Integrated OpenAI and AzureChatOpenAI for language model processing.	+64/-0
audio_trancriber.py `Add audio transcriber tool for YouTube videos` cookbook/Podcast_summarizer_Agents/Tools/audio_trancriber.py Added `audio_transcriber_tool` to extract and transcribe audio from YouTube videos. Utilized `pytube` for downloading audio and `whisper` for transcription.	+24/-0
composio_slack.py `Integrate Slack messaging tool using ComposioToolSet` cookbook/Podcast_summarizer_Agents/Tools/composio_slack.py Added `composio_slack_tool` to integrate Slack messaging using ComposioToolSet.	+5/-0
main.py `Create Streamlit app for podcast summarization` cookbook/Podcast_summarizer_Agents/main.py Created Streamlit app for user input and triggering podcast summarization. Integrated `PodSumCrew` for processing inputs and displaying results.	+21/-0

Documentation

2 files

LICENSE `Add MIT License` cookbook/Podcast_summarizer_Agents/LICENSE - Added MIT License for the project.	+21/-0
README.md `Add README with project description and setup instructions` cookbook/Podcast_summarizer_Agents/README.md Added project description and setup instructions. Included images and sample run video.	+43/-0

Configuration changes

2 files

agents.yaml `Configure agents for summarization and messaging` cookbook/Podcast_summarizer_Agents/config/agents.yaml - Configured agents for podcast summarization and Slack messaging.	+15/-0
tasks.yaml `Configure tasks for summarization and messaging` cookbook/Podcast_summarizer_Agents/config/tasks.yaml Configured tasks for summarizing podcasts and sending messages to Slack.	+11/-0

Dependencies

1 files

requirements.txt

Add project dependencies

cookbook/Podcast_summarizer_Agents/requirements.txt - Added required dependencies for the project.

+7/-0

💡 PR-Agent usage: Comment /help on the PR to get a list of all available PR-Agent tools and their descriptions

codiumai-pr-agent-pro[bot] commented 1 week ago

PR Reviewer Guide 🔍

⏱️ Estimated effort to review [1-5]	4
🧪 Relevant tests	No
🔒 Security concerns	No
⚡ Key issues to review	Possible Bug: The `PodSumCrew` class does not define `self.agents` or `self.tasks` which are referenced in the `crew` method. This might cause runtime errors when trying to create a `Crew` instance.
	Performance Concern: The transcription and summarization process might be resource-intensive and could benefit from asynchronous execution or other performance optimizations.
	Error Handling: There is no error handling for failures in the transcription, summarization, or Slack messaging processes.

codiumai-pr-agent-pro[bot] commented 1 week ago

PR Code Suggestions ✨

Category	Suggestion	Score
Maintainability	Move the instantiation of `AzureChatOpenAI` into an `__init__` method to ensure proper loading of environment variables and manage instantiation timing ___ The instantiation of `AzureChatOpenAI` is done directly within the class attribute definition. This could lead to issues with environment variables not being loaded in time or errors during the class definition phase. It's recommended to move this to a method or inside an `__init__` method. [cookbook/Podcast_summarizer_Agents/Podcast_Summarizer_AI_Agent.py [18-21]](https://github.com/ComposioHQ/composio/pull/240/files#diff-ef19ce3b5c5e128db0b50564646c2005f333fb2aadabd4ce7a5fb79e91b40da2R18-R21) ```diff -llm_model = AzureChatOpenAI(openai_api_version=os.getenv("AZURE_OPENAI_VERSION", "2023-07-01-preview"), - azure_deployment=os.getenv("AZURE_OPENAI_DEPLOYMENT", "gpt4chat"), - azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT", "https://gpt-4-trails.openai.azure.com/"), - api_key=os.getenv("AZURE_OPENAI_KEY")) +def __init__(self): + self.llm_model = AzureChatOpenAI( + openai_api_version=os.getenv("AZURE_OPENAI_VERSION", "2023-07-01-preview"), + azure_deployment=os.getenv("AZURE_OPENAI_DEPLOYMENT", "gpt4chat"), + azure_endpoint=os.getenv("AZURE_OPENAI_ENDPOINT", "https://gpt-4-trails.openai.azure.com/"), + api_key=os.getenv("AZURE_OPENAI_KEY") + ) ``` - [ ] Apply this suggestion Suggestion importance[1-10]: 9 Why: This suggestion significantly improves maintainability and ensures that environment variables are properly loaded before instantiation, which can prevent runtime errors.	9
Best practice	Use the `get` method for dictionary access to avoid potential KeyErrors ___ The `agents_config` dictionary keys are accessed directly which could raise a KeyError if the keys do not exist. It's safer to use the `get` method which will return `None` if the key is not found, or provide a default value. [cookbook/Podcast_summarizer_Agents/Podcast_Summarizer_AI_Agent.py [26]](https://github.com/ComposioHQ/composio/pull/240/files#diff-ef19ce3b5c5e128db0b50564646c2005f333fb2aadabd4ce7a5fb79e91b40da2R26-R26) ```diff -config = self.agents_config['Transcriber_summarizer'] +config = self.agents_config.get('Transcriber_summarizer', {}) ``` - [ ] Apply this suggestion Suggestion importance[1-10]: 8 Why: This suggestion enhances the robustness of the code by preventing potential `KeyError` exceptions, making the code more fault-tolerant.	8
Performance	Cache the `slack_tool` by instantiating it once during class initialization to improve efficiency ___ The `slack_tool` is being instantiated every time the `slack_agent` method is called, which could be inefficient if the method is called multiple times. Consider caching the result or instantiating it once during class initialization. [cookbook/Podcast_summarizer_Agents/Podcast_Summarizer_AI_Agent.py [16]](https://github.com/ComposioHQ/composio/pull/240/files#diff-ef19ce3b5c5e128db0b50564646c2005f333fb2aadabd4ce7a5fb79e91b40da2R16-R16) ```diff -slack_tool = composio_slack_tool() +def __init__(self): + self.slack_tool = composio_slack_tool() ``` - [ ] Apply this suggestion Suggestion importance[1-10]: 8 Why: This suggestion enhances performance by avoiding repeated instantiation of `slack_tool`, which can be inefficient if the method is called multiple times.	8
Possible issue	Convert the `audio_tool` attribute from a list to a single instance if only one tool is required ___ The `audio_tool` attribute is defined as a list containing a single tool, which might be a mistake if the intention was to assign a single tool. If only one tool is required, it should not be wrapped in a list. [cookbook/Podcast_summarizer_Agents/Podcast_Summarizer_AI_Agent.py [15]](https://github.com/ComposioHQ/composio/pull/240/files#diff-ef19ce3b5c5e128db0b50564646c2005f333fb2aadabd4ce7a5fb79e91b40da2R15-R15) ```diff -audio_tool = [audio_transcriber_tool] +audio_tool = audio_transcriber_tool ``` - [ ] Apply this suggestion Suggestion importance[1-10]: 7 Why: This suggestion improves code clarity by ensuring that `audio_tool` is correctly represented as a single instance rather than a list, which is more appropriate given the context.	7

ComposioHQ / composio