🚀 GitHub AI Agent

What is GitHub AI Agent?

GitHub AI Agent is an LLM-powered agent built using the LangGraph ReAct Agent workflow and MCP tools. It integrates with the GitHub API through a dedicated GitHub MCP server, supporting both ACP and A2A protocols for external client communication. Designed for developers and teams, it uses Azure OpenAI GPT-4 as its LLM backend.

How to use GitHub AI Agent?

Configure a .env file with Azure OpenAI credentials and a GitHub personal access token, then run the agent via Docker (docker pull ghcr.io/cnoe-io/agent-github:a2a-latest && docker run ...) or locally (make run-a2a). Interact with the agent using the agent-chat-cli client (e.g., uvx https://github.com/cnoe-io/agent-chat-cli.git a2a).

Key features of GitHub AI Agent

• LangGraph ReAct Agent with MCP tool integration
• Azure OpenAI GPT-4 as the LLM backend
• Multi-protocol support: ACP and A2A
• GitHub API via a dedicated first-party MCP server
• Comprehensive GitHub API support (repos, issues, PRs, branches, commits, projects, teams)
• Docker-based deployment for easy setup

Use cases of GitHub AI Agent

• List and manage your GitHub repositories conversationally
• Create and manage issues with labels and assignments
• Handle pull request operations through natural language
• Perform branch and commit management tasks
• Coordinate team collaboration and project management

FAQ from GitHub AI Agent

What protocols does GitHub AI Agent support?

It supports both the AGNTCY ACP protocol and Google’s A2A protocol for flexible integration with external user clients and multi-agent orchestration.

What are the runtime dependencies?

Python 3.13+, Poetry 2.1.1+, Docker, and access to Azure OpenAI (GPT-4 deployment) with a valid API key and endpoint.

Where does the agent’s data live?

All data operations go through the GitHub API using the provided personal access token. No persistent local storage is mentioned; the agent interacts with GitHub in real time.

What authentication is required?

You need a GitHub classic personal access token with permissions for repo, workflow, admin:org, and other scopes listed in the README, plus an Azure OpenAI API key and endpoint.

How is the GitHub MCP server generated?

The MCP server is generated by the first-party openapi-mcp-codegen utility, ensuring version/API compatibility and software supply chain integrity.