Airbyte Agent Engine installs statistics

Airbyte Agent Engine is the data layer that enables AI agents with real-time access to business data and tools. Connect your AI agents to GitHub and 40+ other apps through secure OAuth. Read and write issues, PRs, repos, code, and more, all through a single platform that handles credentials, token refresh, and data caching for you.

Capabilities

• Read and write GitHub issues, pull requests, repositories, code, and more • Secure OAuth authentication so your end users never share API keys • Built-in credential management with automatic token refresh • Entity caching via Context Store for fast, low-latency search • Works as a standalone Python package, MCP server, or hosted cloud service

Benefits

• Secure access: End users authorize via OAuth instead of copying API keys. Airbyte handles token lifecycle automatically. • Ship faster: Pre-built, type-safe connectors with documented entities and actions. No need to build API integrations from scratch. • Works with any agent framework: Compatible with Pydantic AI, LangChain, FastMCP, and any framework that supports tool calling or MCP.

Getting Started

Requirements:

• Plan: Free for usage of 5k tokens per month. Paid subscription for context store and hosted credential management. • User Permissions: GitHub account with access to the repositories you want to connect. • Availability: Generally available.

Setup Process:

• Install the connector: uv pip install airbyte-agent-gihub • Authenticate with OAuth or a personal access token • Register the connector as a tool in your agent framework and start prompting

Example Prompts

Try these with the Airbyte Agent Engine GitHub connector:

• "List all open issues in my repository" • "Show me pull requests merged this week" • "Get the README for this repo" • "Find commits by a specific author in the last 30 days" • "What branches exist in this repository?"