Scrum Agent

1. Introduction

Scrum Agent is an open-source, AI-powered supervisor designed to enhance agile project management within Discord communities. Acting as a virtual Scrum Master, this agent integrates multiple tools to streamline sprint planning, issue tracking, research, and team collaboration.

By mapping individual Discord channels to corresponding Taiga projects, Scrum Agent facilitates seamless user story management through threaded discussions—ensuring efficient workflow management and improved project oversight.

Join our Discord Server to test the Scrum Agent and explore its features.

Project Goals, Audience & Maturity

Scrum Agent is particularly helpful for small dev teams without a dedicated Scrum Master, but can also be used by larger teams looking to keep their Discord conversations and Taiga board in sync. Currently, it’s designed for Discord and Taiga, but we plan to extend support to Slack and Jira as well.

The project is still in an early stage and is under active development—use it at your own risk. That said, we welcome contributions and feedback from the community to help shape its future.

2. Features

Scrum Agent offers a range of capabilities to support agile project management:

Discord Management

Seamless Integration:
Acts as a central hub for project updates and agile workflow discussions.
Automated Task Tracking:
Retrieves and organizes sprint-related messages to ensure no detail is overlooked.
Efficient Search & Retrieval:
Enables quick access to relevant messages, posts, and threads.
Channel and Thread Organization:
Provides a structured list of channels and active threads for easy navigation.

Taiga Scrum Master

Issue Management:
Track and update sprints, user stories, and tasks efficiently.
User Story Details:
Retrieve comprehensive information, including history and assigned users.
Task Updates:
Modify descriptions, update statuses, and assign watchers as needed.
User Story Creation:
Enhance team collaboration by creating new user stories directly.

Web Tools

DuckDuckGo:
Perform efficient web searches.
ArXiv:
Access a wide range of research papers.
YouTube & Wikipedia:
Retrieve information quickly for research or reference.
Web Navigation:
Gather additional context and relevant data.

DeepSeek Reasoning

Advanced Problem-Solving:
Leverage AI-driven insights to address complex challenges.
Strategic Analysis:
Enhance decision-making with robust analytical tools.
Abstract Reasoning:
Gain predictive insights and better understand abstract concepts.

3. Installation and Setup

Python Project Setup

Create and activate a Conda environment:

conda create -n scrumagent python=3.11 -y
conda activate scrumagent

Upgrade pip and install dependencies:

pip install --upgrade pip
pip install -r requirements.txt

Install Ollama for DeepSeek Agent:

Download and install Ollama (adjust size as needed):

curl -fsSL https://ollama.com/install.sh | sh
ollama pull deepseek-r1:8b
ollama pull llama3.2-vision:11b

Discord Bot Setup

1. Create Your Discord Bot

Access the Developer Portal:
Visit the Discord Developer Portal and sign in with your Discord account.
Create a New Application:
Click "New Application", give your bot a name, and create the application.

2. Configure OAuth2 Settings

Navigate to the OAuth2 Tab:
Once your application is created, click on the OAuth2 tab.
Enable the Bot Option:
In the OAuth2 settings, activate the "bot" option.
Set Bot Permissions:
Under the OAuth2 Bot Permissions section, ensure that the following permissions are enabled:
- View Channels
- Send Messages
- Create Public Threads
- Create Private Threads
- Manage Threads
- Send Messages in Threads
- Manage Messages
- Read Message History
- Add Reactions
Enable Intents:
Go to the "Bot" tab and enable:
- Server Members Intent
- Message Content Intent

3. Add the Bot to Your Server

Generate the OAuth2 URL:
After configuring the permissions, generate the OAuth2 URL.
Invite the Bot:
Use the generated URL to invite the bot to your Discord server, and follow the on-screen instructions to complete the process.

4. Retrieve Your Bot Token

Go to the "Bot" Tab:
In your application settings, click on the "Bot" tab in the sidebar.
Reveal the Token:
Under the "Build-A-Bot" section, locate the "Token" field. Click the "Reset Token" or "View Token" button to reveal your bot token.

Important: Never share your bot token publicly or commit it to version control. Treat it like a password.
Copy Your Token:
Copy the bot token for the next step.
Add the Token to Your .env File:
- If you haven’t created a .env file, copy .env.example to .env.
- Add the following line to your .env file:
```
DISCORD_TOKEN=YOUR_DISCORD_BOT_TOKEN_HERE
```

Environment Variables

Set Up Environment Variables:
- Copy the file .env.example and rename it to .env.
- Open the new .env file and fill in the required details.
Customize the Mapping:
- Copy config/taiga_discord_maps.yaml.example and rename it to config/taiga_discord_maps.yaml.
- Edit this file as needed to match your project settings.
Specific Settings:
- MAX_MSG_MODE: 'trim' (only keep the last MAX_MSG_COUNT messages) or 'summary' (when the message count exceeds MAX_MSG_COUNT, summarize the messages and keep the summary as context).

4. Usage Examples

Below are examples showing how to create and update Taiga issues or user stories directly from Discord. Changes made in Discord appear in Taiga, making project management faster and easier.

Updating a User Story

Left: Updating a user story in Discord
Right: The updated user story in Taiga

Creating or Updating an Issue

Left: Creating or updating an issue in Discord
Right: The newly created or updated issue in Taiga

Try It Out!

Join our Discord Server to test the Scrum Agent and explore its features.

5. System Architecture

Overview

Scrum Agent is built on the Langgraphs Supervisor Agent, which allows for flexible integration of multiple agents and tools. For more technical details, please refer to the Langgraphs documentation.

Architecture Diagram

Figure: Architecture diagram of Scrum Agent's multi-agent setup.

Adding Agents and Tools

Adding Agents:
Insert new agent nodes in build_agent_graph.py.
Supervisor Configuration:
Update supervisor_agent.py with the agent's name and description. Ensure that the names are consistent.
Tool Integration:
Follow the Langgraphs Documentation for guidance on adding and integrating additional tools.

6. Tracing with LangSmith

To enable tracing with LangSmith, follow these steps:

Set Environment Variables:
```
LANGCHAIN_TRACING_V2=True
LANGCHAIN_API_KEY=your_api_key_here
LANGCHAIN_PROJECT=your_project_name
```
Replace your_api_key_here with your actual API key (refer to .env.example for guidance).
Learn More:
- Tracing Documentation
- LangSmith Website

This configuration allows you to monitor and trace your application's activity using LangSmith.

7. Testing

If you want to use unit tests or add test files. You can go to the tests folders.The "agent_tests" folder belongs to the agents, and tool tests belong to the tools. For test mocking, you need to use the Taiga API. You can change from .env file

7.1 Local deployment

you can use taiga-docker (https://github.com/taigaio/taiga-docker)

TAIGA_URL="http://localhost:9000/"

7.2 Taiga io

TAIGA_URL="https://api.taiga.io" https://api.taiga.io

7.3 running tests

-Example usage

python3 testfile.py

8. Roadmap

See the Taiga Project for the latest updates and planned features.

9. Contact

For inquiries, support, or contributions, please open an issue here or at Taiga
or join our Discord or contact [email protected]

Name		Name	Last commit message	Last commit date
Latest commit History 78 Commits
assets		assets
config		config
scrumagent		scrumagent
tests		tests
.env.example		.env.example
.gitignore		.gitignore
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
LICENSE		LICENSE
README.md		README.md
langgraph.json		langgraph.json
requirements.txt		requirements.txt

License

Shikenso-Analytics/ScrumAgent

Folders and files

Latest commit

History

Repository files navigation