Every agent you build eventually needs to talk to the outside world. Your customer support agent needs access to Zendesk tickets. Your sales copilot needs to query open Salesforce opportunities. Your engineering assistant needs to create GitHub issues. Agents often need to search and discover the data sources relevant to their queries as well, which may stretch across multiple data sources and SaaS tools.
Before you know it, you're managing OAuth flows, pagination, rate limits, and API translations for every data source, building a mini integration platform instead of your actual product.
The natural instinct is to reach for off-the-shelf MCP servers or tool connectors. These work fine in a demo. But in production, they flood your agent's context window with raw API responses, expose PII to the model, and give you no way to enrich or structure the data before it reaches your agent. The result is slow, inaccurate responses that can't reason across systems. You don't just need a way to call external APIs. You need infrastructure that gives your agent fast, structured access to business context.
The result: most teams either limit their agents to a handful of sources, or spend more time on integration plumbing than on the agent logic that actually matters. What should be a solved problem (calling an API) becomes a recurring tax on every feature you ship.
Introducing Airbyte Agent Engine In December, we announced agent connectors , open source Python connectors purpose-built to give AI agents consistent, real-time access to operational data for context-rich decision making.
Today we’re excited to announce the Airbyte Engine is officially in public beta!
The Airbyte Agent Engine builds on the agent connectors to give your agents a unified interface to external data sources. Our platform includes a fully-managed authentication module supporting OAuth, hosted agent connectors, and the Context Store (more below). This ensures that your agent only requests data from external systems when absolutely necessary, optimizing processes for compute and latency.. We're launching with 20+ agent connectors today, including HubSpot, Salesforce, Gong, Linear, GitHub, and more, with hundreds more coming fast from Airbyte's native ecosystem of 700+ data connectors .
Connecting your AI agents to core business applications (Salesforce, Slack, HubSpot, etc.) with real-time fetch and write capabilities takes just ~10 lines of code. The following is an example for integrating end user GitHub repositories to your agents:
Here’s how easy it is to implement two of our connectors in PydanticAI:
gong = GongConnector(auth_config=AirbyteAuthConfig(...))
hubspot = HubSpotConnector(auth_config=AirbyteAuthConfig(...))
@agent.tool_plain
@gong.tool_utils
async def gong_execute (entity, action, params ):
return await gong.execute(entity, action, params or {})
response = await agent.run(
"Find my latest Gong call and create a new "
"opportunity in HubSpot with the key details."
)The Context Store Complex natural language queries requiring searching across lists of records, such as "list all customers closing this month with deal sizes greater than $5000", typically require substantial work: sequences of paginated API calls, and filtering across large datasets. This often results in unbounded context window growth and rate limit issues, both of which lead to perceived downtime and broken trust with customers.
The Context Store stores a subset of relevant data from connected sources in Airbyte-managed context storage, enabling your agents to perform efficient search operations without repeatedly querying vendor APIs. The Context Store helps improve latency of your more complex prompts by allowing your agents to reliably search across records in half a second or less.
When enabled, the Context Store automatically populates during initial setup and refreshes hourly. Each connected source gets its own isolated data store with organization-level access control.
This is a true game changer for agentic data infrastructure. The Context Store uses data replication architecture that Airbyte has built over the last five years, enabling search at a level no other competitor can provide. Replication powered search is absolutely critical for building agentic products at scale, and the Airbyte Agent Engine is the first agent infrastructure product to deliver it.
Getting Started with the Airbyte Agent Engine Create an account and login here .
Once you are logged in, you can access the onboarding flow which will help you set up your first connector. In case you wish to do it manually, or need help, you can follow the steps below.
You have two options for connecting data sources:
Option 1: Use our authentication module. Our Agent Engine provides an out-of-the-box auth module to ease integrating with and managing credentials from outside systems. The auth module makes it easy to add these integrations to your app without building auth flows from scratch.Option 2: Register credentials directly via API. If you prefer to provide your own integrations page and manage credentials yourself, you can register them with us directly through the API.Once you’ve registered a connector, you can immediately access its data by integrating our libraries into your agent. Rather than exposing dozens of tools per API (one for each endpoint) for integration, we generally recommend starting by implementing a single tool for each connector. Our @connector.tool_utils decorator helps manage tool sprawl. With it, your agent gets the flexibility to query any entity and action, while your tool count stays manageable. You can also script these same operations through the Agent CLI when you want your agent to drive the platform directly from the command line.
In a typical github_execute tool, you have full control over the response from the API. This flexibility allows you to:
Manage the response schema to fit your agent's context windowRoll custom error handling for graceful degradationMask PII before it reaches your agentEnrich the response with data from other toolsCode snippets to get started, authenticate, and execute connectors are in our documentation (example for Github ). For a higher level view of how to build an agentic app from scratch using Airbyte, you can follow this quick guide .If you want more guidance, talk to our team!
Help Us Build What You Need Agent Engine is in public beta, which means we're shipping fast and building in the open. We're adding new connectors every week, and your feedback directly shapes what comes next.
Here's what we want to hear from you:
What connectors do you need? We're launching with 20+ today, but our roadmap is driven by demand. If your agent needs access to a tool we don't support yet, tell us and we'll prioritize it.
What's working? What isn't? Whether it's the Context Store, the auth module, or the developer experience, we want to know what's helping you ship faster and what's getting in your way.
What features would make Agent Engine indispensable? We have a packed roadmap, but the best ideas come from builders working on real problems.
Share your feedback directly in our Slack community , open a feature request on GitHub , or talk to our team . We read everything.
Start building today at app.airbyte.ai .
