Project Setup Guide

This guide will help you set up and run the LangGraph PM Maestro project locally.

Clone the Repository

git clone https://github.com/PrasanKumar93/langgraph-pm-maestro.git
cd langgraph-pm-maestro

Install Dependencies

npm install

Configure Environment Variables

Copy the example environment file to create your own configuration:

cp .env.example .env

Open the .env file and provide the necessary credentials
- Mandatory fields
  - DEFAULT_LLM_PROVIDER : openai (openai/ aws_bedrock)
  - If OpenAI is selected as DEFAULT_LLM_PROVIDER, then the following fields are required
    - OPENAI_API_KEY : For LLM provider OpenAI
  - If AWS Bedrock is selected as DEFAULT_LLM_PROVIDER, then the following fields are required
    - AWS_BEDROCK_MODEL_NAME : For LLM provider AWS Bedrock
    - AWS_REGION : For AWS region
    - AWS_ACCESS_KEY_ID : For AWS access key id
    - AWS_SECRET_ACCESS_KEY : For AWS secret access key
    - AWS_SESSION_TOKEN : For AWS session token
  - REDIS_URL : For checkpointers, vector DB, LLM cache ..etc
  - TAVILY_SEARCH_API_KEY : To search the latest web content for competitor analysis of the requested feature
- Optional fields group
  - SALESFORCE CONFIGURATION : To search the requested feature in Salesforce
  - JIRA CONFIGURATION : To search the requested feature in Jira
  - SLACK CONFIGURATION : To access the agent via slack

How to generate required env variables

Running the application

Running the agent locally in browser (Langgraph Studio)

npm run dev

Launches the agent locally and opens the studio interface in your browser at: https://smith.langchain.com/studio?baseUrl=http://localhost:2024 (Test in chrome browser)

Running the Slack Bot

To interact with the agent through Slack:

npm run start-slack-bot

Once running, you can communicate with the bot in your registered Slack channel. The agent will respond to your messages automatically.